1
00:00:00,140 --> 00:00:03,020
I'm Miko Pawlikowski
and this is HockeyStick.

2
00:00:06,580 --> 00:00:10,870
Today we're talking about robo advisors,
a topic in between tech and finance.

3
00:00:11,040 --> 00:00:14,799
To do that I'm bringing in some heavy
hitters, the authors of "Build a Robo

4
00:00:14,799 --> 00:00:19,689
Advisor with Python from Scratch",
Rob Reider and Aleksander Michalka.

5
00:00:20,050 --> 00:00:24,630
Rob has been a quantitative hedge fund
portfolio manager for over 15 years.

6
00:00:24,690 --> 00:00:29,040
He holds a PhD in finance from the Wharton
School and is an adjunct professor at

7
00:00:29,119 --> 00:00:33,679
NYU, where he teaches a graduate course
in the math finance department called Time

8
00:00:33,680 --> 00:00:36,230
Series Analysis and Statistical Arbitrage.

9
00:00:36,610 --> 00:00:39,359
Alex leads the investment
research group at Wealthfront.

10
00:00:39,820 --> 00:00:44,390
He holds a BA in Applied Mathematics
from UC Berkeley and a PhD in Operations

11
00:00:44,390 --> 00:00:46,010
Research from Columbia University.

12
00:00:46,825 --> 00:00:50,155
Welcome to this episode, and
thank you for flying HockeyStick.

13
00:00:50,973 --> 00:00:52,933
All right, Rob, we're
going to start with you.

14
00:00:53,213 --> 00:00:55,853
when I read your resume,
I felt very small,

15
00:00:56,083 --> 00:01:00,243
not only to mention your
extensive experience, but also

16
00:01:00,253 --> 00:01:01,683
all the right names on it.

17
00:01:01,723 --> 00:01:06,863
The Wharton, a PhD in finance,
NYU, adjunct professor.

18
00:01:07,233 --> 00:01:10,253
Can you tell us a little bit about what
you're teaching and what you're doing?

19
00:01:10,963 --> 00:01:13,283
after my PhD, I, worked at a bank.

20
00:01:13,283 --> 00:01:17,643
I worked on, derivative securities, then I
moved to their proprietary trading group.

21
00:01:17,723 --> 00:01:21,943
then, I moved to, an actual hedge fund,
Millennium Partners, big hedge fund.

22
00:01:22,483 --> 00:01:24,703
went to another hedge fund, which blew up.

23
00:01:24,803 --> 00:01:25,813
it's a very sad story.

24
00:01:25,813 --> 00:01:29,683
It was started by a medical doctor
and, wanted to, But then somebody

25
00:01:29,683 --> 00:01:33,203
on the medical side, the health care
side, committed insider trading.

26
00:01:33,213 --> 00:01:34,733
It was a front page of
Wall Street Journal.

27
00:01:34,813 --> 00:01:37,113
and the hedge fund basically, blew up.

28
00:01:37,483 --> 00:01:40,093
but then after that I had a
very interesting, experience.

29
00:01:40,093 --> 00:01:43,053
I worked at a wonderful
company called Quantopian.

30
00:01:43,239 --> 00:01:46,249
a backtesting platform in Python.

31
00:01:46,269 --> 00:01:51,999
So you can go into their platform,
backtest different trading strategies.

32
00:01:51,999 --> 00:01:53,029
They supply data.

33
00:01:53,039 --> 00:01:54,899
They had a lot of built in code.

34
00:01:55,329 --> 00:01:58,734
and then their basic business model
was they were going to try to be

35
00:01:58,734 --> 00:02:00,374
the first crowdsourced hedge fund.

36
00:02:00,374 --> 00:02:04,074
So if you thought you had a good
strategy, you could just hit a button

37
00:02:04,104 --> 00:02:07,394
and be considered for actual money.

38
00:02:07,404 --> 00:02:11,054
And they actually got an investor
to see their hedge funds.

39
00:02:11,434 --> 00:02:13,654
and it was interesting for me
because I started my life as an

40
00:02:13,654 --> 00:02:15,104
engineer and did a lot of coding.

41
00:02:15,104 --> 00:02:17,944
And then I moved away
from coding for years.

42
00:02:17,944 --> 00:02:20,884
Like when I worked at millennium,
I had people work for me and

43
00:02:20,884 --> 00:02:21,974
they would do all the coding.

44
00:02:22,489 --> 00:02:27,679
So my coding skills atrophied, but when
I, went back into Quantopia and I had

45
00:02:27,689 --> 00:02:31,629
to learn Python, like I'm so old that
I learned Fortran as my first language.

46
00:02:31,679 --> 00:02:34,479
I picked up a Python book
and I just loved Python.

47
00:02:34,589 --> 00:02:37,359
so that sparked my
interest, in doing Python.

48
00:02:37,849 --> 00:02:41,579
and then I actually had A stint
at a robo advisor as well, which

49
00:02:41,579 --> 00:02:43,429
was a startup robo advisor.

50
00:02:43,429 --> 00:02:45,979
I think it's very hard to be a startup
anything, especially when you're

51
00:02:45,979 --> 00:02:50,359
competing against large incumbents like,
Vanguard and Schwab and things like that.

52
00:02:50,369 --> 00:02:53,889
and then I teach at NYU, so I've
been teaching for, over 15 years.

53
00:02:53,979 --> 00:02:58,569
I teach a course on quantitative
trading for a master's program,

54
00:02:58,599 --> 00:03:01,779
it's a one and a half year
master's in mathematical finance.

55
00:03:03,439 --> 00:03:07,219
then you also have that online course
time series analysis in Python.

56
00:03:07,219 --> 00:03:09,289
So I guess you, taught yourself Python.

57
00:03:09,339 --> 00:03:10,909
now you're teaching everybody else.

58
00:03:11,069 --> 00:03:14,739
yeah, so when I was at Quantopian,
there's this company called Datacamp.

59
00:03:14,799 --> 00:03:16,119
Quantopian was in Boston.

60
00:03:16,179 --> 00:03:17,859
Datacamp used to be in Boston,
now they're in New York.

61
00:03:17,909 --> 00:03:19,859
I guess their offices were
near Quantopians, and they

62
00:03:19,859 --> 00:03:20,869
were just starting out.

63
00:03:21,259 --> 00:03:26,389
And they said: we need somebody to teach a
course in time series analysis in Python.

64
00:03:26,399 --> 00:03:29,579
So Quantopian suggested that I do it.

65
00:03:29,609 --> 00:03:35,024
And that's actually related to our,
origin story, because once I did that

66
00:03:35,034 --> 00:03:39,914
online course, which became fairly
popular, a few publishers reached out

67
00:03:39,914 --> 00:03:43,524
to me and said, do you want to do a
book on time series analysis in Python?

68
00:03:43,524 --> 00:03:47,534
And at the time I was busy, I was bored
of the subject, so I turned it down.

69
00:03:48,024 --> 00:03:51,544
the publisher Manning said, Do you,
we pick somebody else to do it.

70
00:03:51,554 --> 00:03:52,984
Do you want to edit what they're doing?

71
00:03:53,404 --> 00:03:55,854
and then, I started like
editing somebody else's work.

72
00:03:55,854 --> 00:03:58,294
And so like, why am I spending all this
time, editing somebody else's work?

73
00:03:58,294 --> 00:04:00,084
Maybe I should just write my own book.

74
00:04:00,084 --> 00:04:03,934
And the next kind of offer that came
along was this book on robo advising and

75
00:04:03,934 --> 00:04:05,334
Python, which I know something about.

76
00:04:06,674 --> 00:04:10,844
And then I knew I know stuff
takes me a long time to do.

77
00:04:10,844 --> 00:04:11,964
I'm a perfectionist.

78
00:04:11,974 --> 00:04:15,204
Things take me three times longer
than it would take somebody else.

79
00:04:15,534 --> 00:04:17,454
So I knew it would take me
forever to write a book.

80
00:04:17,454 --> 00:04:19,274
So I knew I had to find a co author.

81
00:04:19,534 --> 00:04:22,279
And I basically just cold called Alex.

82
00:04:22,279 --> 00:04:23,279
I didn't know who he was.

83
00:04:23,279 --> 00:04:26,439
I just saw that he was head
of research at a Wealthfront.

84
00:04:26,439 --> 00:04:30,329
He had the perfect background,
a PhD, worked at a robo advisor.

85
00:04:30,329 --> 00:04:33,969
So I just reached out to him and
said, would you be interested?

86
00:04:33,969 --> 00:04:37,289
And we hit it off and that's
how we started collaborating.

87
00:04:37,304 --> 00:04:38,404
I

88
00:04:38,509 --> 00:04:39,989
So you were roped in Alex.

89
00:04:40,049 --> 00:04:40,779
Is that right?

90
00:04:40,839 --> 00:04:42,679
One day you just wake up, you get a call.

91
00:04:42,679 --> 00:04:44,129
It's Oh, fine.

92
00:04:44,129 --> 00:04:45,489
I'll write a book with you.

93
00:04:45,499 --> 00:04:47,909
yeah, I had kind of always
wanted to write a book.

94
00:04:47,969 --> 00:04:52,849
never was sure what the topic would
be, had a few candidates in mind.

95
00:04:52,879 --> 00:04:56,439
But then, when I met Rob, I thought
it was the perfect fit, like you

96
00:04:56,439 --> 00:04:59,719
mentioned, I've been working at
Wealthfront, as the head of research

97
00:04:59,719 --> 00:05:01,319
for about two years at that point.

98
00:05:02,229 --> 00:05:04,859
And, Rob had basically
already written the book.

99
00:05:04,979 --> 00:05:08,679
He had a full outline of all the chapters
and the sub chapters and all this stuff.

100
00:05:08,689 --> 00:05:11,199
So it made it really
easy for me to say yes.

101
00:05:11,739 --> 00:05:16,639
I also saw that you worked at weather bill
to do weather derivative pricing models.

102
00:05:17,139 --> 00:05:18,559
that sounds very interesting to me.

103
00:05:18,559 --> 00:05:22,839
can you talk a little bit about what
that actually means in practice?

104
00:05:23,209 --> 00:05:23,759
Yeah.

105
00:05:23,779 --> 00:05:26,599
So the company was started, in 2006.

106
00:05:26,599 --> 00:05:28,829
It was a friend that I knew from college.

107
00:05:28,929 --> 00:05:30,549
and he roped me into it.

108
00:05:30,599 --> 00:05:33,799
To use that term, I was going to go
straight through to grad school, but right

109
00:05:33,799 --> 00:05:37,399
before I was graduating, he convinced me
to come on and work for him for a year.

110
00:05:38,619 --> 00:05:41,369
And the weather derivative thing
isn't the most important part.

111
00:05:42,109 --> 00:05:48,459
really what we were trying to do is build
a, basically an insurance solution for

112
00:05:48,569 --> 00:05:50,649
businesses that were affected by weather.

113
00:05:50,954 --> 00:05:56,614
So you can think of things like golf
courses or events, if you're having a

114
00:05:56,614 --> 00:06:01,914
big outdoor concert or sports event and
expecting to sell a lot of tickets, sell

115
00:06:01,914 --> 00:06:05,194
a lot of merchandise and it gets rained
out, you could lose a lot of money.

116
00:06:05,654 --> 00:06:10,804
Or if there's a, really warm
winter and you're a power utility,

117
00:06:10,824 --> 00:06:13,134
you might not sell as much gas.

118
00:06:13,529 --> 00:06:19,349
so you could use a weather derivative
contract to, hedge against whatever

119
00:06:19,649 --> 00:06:21,059
weather peril you're facing.

120
00:06:21,909 --> 00:06:24,929
the market had been mostly in that kind
of utility sector that I mentioned.

121
00:06:25,949 --> 00:06:28,769
for the longest time, but it
was a very manual process.

122
00:06:28,789 --> 00:06:33,009
If you wanted to hedge weather risk,
you had to get your team of lawyers and

123
00:06:33,009 --> 00:06:38,389
go to a bank and talk to their team of
lawyers and get the details hammered out.

124
00:06:39,109 --> 00:06:46,229
what we did was use technology to make the
process a lot simpler, cheaper, faster.

125
00:06:47,009 --> 00:06:49,649
And, let's see, it was all online.

126
00:06:49,649 --> 00:06:51,779
So you never actually had to talk
to anybody could go online and

127
00:06:51,779 --> 00:06:54,379
configure a contract and just
buy it right there on the spot.

128
00:06:55,379 --> 00:06:59,049
So how much of that was
actual weather modeling?

129
00:06:59,799 --> 00:07:00,769
There was a good amount.

130
00:07:00,769 --> 00:07:04,009
So there are multiple ways
to price a weather contract.

131
00:07:04,799 --> 00:07:07,839
you can buy historical weather
data, which is what we did.

132
00:07:08,749 --> 00:07:13,139
And, the simplest way is just model
the payouts that would have happened.

133
00:07:13,684 --> 00:07:19,794
historically and figure out a fair
value for the contract based on that.

134
00:07:20,674 --> 00:07:26,394
Another way that you can do it is actually
simulate the underlying weather processes,

135
00:07:26,974 --> 00:07:31,134
like the daily temperature values and the
daily precipitation values, and then run

136
00:07:31,134 --> 00:07:36,694
lots and lots of simulations and, take the
average compute payouts for the contract

137
00:07:36,694 --> 00:07:41,204
based on those weather simulations, and
then, come up with a price based on that.

138
00:07:42,009 --> 00:07:46,099
So we started with the easier method,
that was my first task on the job,

139
00:07:46,479 --> 00:07:50,339
and then later we started moving into
the more complex methods where you're

140
00:07:50,339 --> 00:07:55,869
actually modeling the underlying, weather
processes, which is a lot of work,

141
00:07:55,889 --> 00:08:01,539
especially when you have thousands of,
locations, which are usually airports, but

142
00:08:01,539 --> 00:08:06,299
not always, thousands of locations across
the US, across Europe, Canada, pretty

143
00:08:06,299 --> 00:08:07,759
large scale, pretty difficult problem.

144
00:08:08,309 --> 00:08:11,739
So for anybody listening to this,
you'll probably notice that it's

145
00:08:11,739 --> 00:08:15,479
not our typical technical, episode.

146
00:08:15,489 --> 00:08:20,359
We're going to expand our horizons a
little bit and try to understand an entire

147
00:08:20,359 --> 00:08:22,829
new environment, the finance environment.

148
00:08:22,879 --> 00:08:26,709
But before we jump into the robot
advisors, I want to use this opportunity

149
00:08:26,709 --> 00:08:31,949
because I don't usually get people
with, 15 years, experience in a hedge

150
00:08:31,949 --> 00:08:37,069
funds portfolio manager role to a lot of
people listening to this, they're going

151
00:08:37,069 --> 00:08:40,639
to be coming from software engineering
background and probably their exposure

152
00:08:40,669 --> 00:08:46,674
of how hedge funds actually work is
shaped primarily by things like billions.

153
00:08:46,774 --> 00:08:49,604
how does it compare to real life?

154
00:08:49,704 --> 00:08:53,864
from your experience, are there elements
of billions that you would say are

155
00:08:53,934 --> 00:08:56,344
representative, or is it all Hollywood?

156
00:08:57,574 --> 00:09:01,444
only watched a handful of episodes of
Billions, so I can't speak to that.

157
00:09:01,444 --> 00:09:05,244
But of the few episodes I watched, I
thought it was actually pretty realistic.

158
00:09:05,244 --> 00:09:07,514
And I think the writers, know Wall Street.

159
00:09:07,514 --> 00:09:10,604
this like sort of two
types of hedge funds.

160
00:09:10,624 --> 00:09:14,424
There's, Traders that use
fundamental information.

161
00:09:14,454 --> 00:09:19,494
So they'll, analyze companies, products
and their financial statements, and

162
00:09:19,494 --> 00:09:23,144
they'll listen to earnings calls and
they'll talk to sell side analysts.

163
00:09:23,144 --> 00:09:24,684
And those are like the
fundamental traders.

164
00:09:24,914 --> 00:09:27,754
And then you have the quantitative
traders that don't know, anything

165
00:09:27,754 --> 00:09:28,614
about what the company does.

166
00:09:28,614 --> 00:09:32,694
They just, use statistics and computers
to come up with trading strategies

167
00:09:32,894 --> 00:09:34,644
and in billions, they had both.

168
00:09:34,774 --> 00:09:36,574
and that's millennium was like that too.

169
00:09:36,624 --> 00:09:40,864
Millennium, had something like,
150 different little tiny groups

170
00:09:40,864 --> 00:09:44,094
or silos and each group might
be two, three, four people.

171
00:09:44,384 --> 00:09:47,624
and some of them traded based on
fundamentals and usually focused on

172
00:09:47,714 --> 00:09:52,294
a single sector, because you can only
really be an expert on one little area.

173
00:09:52,544 --> 00:09:56,275
And then they also had a large number
of quantitative traders as well.

174
00:09:56,910 --> 00:09:57,990
and I should say that.

175
00:09:58,595 --> 00:10:02,795
Millennium is considered a multi strategy
hedge fund, but there's lots of other

176
00:10:02,845 --> 00:10:07,085
types of hedge funds, there's some
hedge funds that have large groups,

177
00:10:07,085 --> 00:10:11,135
like there's a known hedge fund,
Two Sigma, that has instead of small

178
00:10:11,135 --> 00:10:14,825
silos of two, three, four people, they
have large groups of a hundred people

179
00:10:14,855 --> 00:10:16,825
and they work in a different way.

180
00:10:16,825 --> 00:10:20,565
They also do quantitative stuff, but
in a large collaborative environment.

181
00:10:20,575 --> 00:10:24,155
Millennium was very secretive, groups
didn't talk to each other, give away

182
00:10:24,155 --> 00:10:28,485
their secret sauce, which solves some
problems and creates other problems.

183
00:10:28,485 --> 00:10:32,745
on one hand, when you work in small
silos, you have to reinvent the wheel.

184
00:10:32,745 --> 00:10:35,275
You have to come up with
your own trading algorithms.

185
00:10:35,295 --> 00:10:38,925
So there's no common algorithms
that everyone could use, but

186
00:10:38,925 --> 00:10:40,875
also it aligns incentives.

187
00:10:40,985 --> 00:10:43,605
You don't have to worry about
people like stealing your stuff

188
00:10:43,605 --> 00:10:44,905
and going to another hedge fund.

189
00:10:44,905 --> 00:10:48,525
And, Also, at Millennium, every
little group pays for their own data.

190
00:10:48,525 --> 00:10:49,965
So when I worked at, J.

191
00:10:49,965 --> 00:10:50,115
P.

192
00:10:50,115 --> 00:10:53,075
Morgan, we subscribe to all
these different data sources.

193
00:10:53,315 --> 00:10:54,735
And we don't even know why we subscribe.

194
00:10:54,735 --> 00:10:57,195
somebody at some point may have
said, Oh, I need this, data.

195
00:10:57,195 --> 00:10:59,395
And 10 years later, we're
still subscribing to it.

196
00:10:59,404 --> 00:11:01,994
Whereas at Millennium, it
comes out of our own P&L.

197
00:11:02,064 --> 00:11:06,205
So we have the incentive to try to,
just, get the data that we need so

198
00:11:06,215 --> 00:11:08,685
it solves a lot of incentive issues.

199
00:11:09,772 --> 00:11:12,622
So there's a lot of them and there's
going to be a lot of variation.

200
00:11:13,072 --> 00:11:17,732
And then is it true to say that in
the recent decades, we've seen this

201
00:11:17,752 --> 00:11:23,472
rise of this new breed of, investment,
funds that we call robo advisors,

202
00:11:23,512 --> 00:11:27,832
like an umbrella term for all of
them, or is it older of an invention?

203
00:11:28,634 --> 00:11:30,945
the hedge fund world is very
different from the robo advisor world.

204
00:11:30,954 --> 00:11:34,275
The hedge fund world tries to
figure out ways to beat the market.

205
00:11:34,545 --> 00:11:38,955
Whereas I would say the robo advisor
world is investing in kind of

206
00:11:39,055 --> 00:11:40,885
low cost index funds in the way.

207
00:11:41,120 --> 00:11:45,050
They try to beat the market is
through, saving money on taxes and

208
00:11:45,060 --> 00:11:46,510
having higher after tax returns.

209
00:11:46,520 --> 00:11:51,690
So robo advisors typically don't
try to, figure out like hedge funds

210
00:11:51,690 --> 00:11:54,650
do, like which stocks to buy, which
stocks are better, which one's going

211
00:11:54,650 --> 00:11:58,220
to outperform the market, what their
hedge funds look for different anomalies

212
00:11:58,230 --> 00:11:59,540
and try to take advantage of that.

213
00:11:59,760 --> 00:12:03,090
Robo advisors take a totally
different approach and say we're

214
00:12:03,090 --> 00:12:05,050
just going to invest in index funds.

215
00:12:05,050 --> 00:12:07,220
We're going to do in a
very tax efficient way

216
00:12:07,840 --> 00:12:11,820
So what was the thing that got you
interested in robo advisors coming

217
00:12:11,820 --> 00:12:15,340
from the background of, of hedge funds
and beating the markets and doing all

218
00:12:15,340 --> 00:12:20,989
this ambitious things to moving on
to, something more for everybody, the

219
00:12:21,140 --> 00:12:23,829
average Joe's, robo advisor thing.

220
00:12:23,849 --> 00:12:26,729
What was the trigger, that
got you interested in that?

221
00:12:27,074 --> 00:12:28,944
even working at hedge funds.

222
00:12:28,944 --> 00:12:33,504
I've always been interested in personal
taxes ways to save money on taxes.

223
00:12:33,824 --> 00:12:39,174
20 years ago, I was telling people
at work about this concept of asset

224
00:12:39,194 --> 00:12:42,614
location, not asset allocation, but
something called asset location, which

225
00:12:42,614 --> 00:12:46,774
we could talk about later, but this
was something that I was espousing

226
00:12:47,084 --> 00:12:48,414
many years before it was popular.

227
00:12:48,414 --> 00:12:51,224
So I've always been interested
in personal finance.

228
00:12:51,225 --> 00:12:53,674
I do my own taxes on TurboTax.

229
00:12:53,674 --> 00:12:57,454
So as a result, I've learned
a lot about the tax code by

230
00:12:57,814 --> 00:12:58,884
being a 'do it yourself' person.

231
00:12:58,884 --> 00:13:02,624
And this idea of trying to, save money
on taxes, for example, is something

232
00:13:02,624 --> 00:13:03,804
that I've always been interested in.

233
00:13:05,109 --> 00:13:08,569
And what about you, Alex, you work
at Wealthfront, which is, one of the

234
00:13:08,814 --> 00:13:13,294
more easily recognizable, at least
from where I'm sitting, robo advisors.

235
00:13:13,294 --> 00:13:16,074
What was the journey for you
to, get interested in that?

236
00:13:16,174 --> 00:13:16,734
Yeah.

237
00:13:16,734 --> 00:13:20,394
So let's see, like I mentioned
before, I started in tech, but

238
00:13:20,394 --> 00:13:23,314
the company that I was working
for had a financial slam to it.

239
00:13:23,404 --> 00:13:26,924
We were risk management company, where
in particular, the risk that we were

240
00:13:26,924 --> 00:13:28,784
helping people manage against was weather.

241
00:13:29,999 --> 00:13:34,969
And then after grad school, I
actually worked at a, hedge fund

242
00:13:34,979 --> 00:13:37,189
also called AQR Capital Management.

243
00:13:37,199 --> 00:13:43,189
It ran a variety of strategies, some
were hedge fund strategies where, you're

244
00:13:43,299 --> 00:13:46,476
going along a bunch of stocks and short
a bunch of stocks and, trying to, Hedge

245
00:13:46,476 --> 00:13:50,216
out market risk, make a return and
somewhere along only strategies where,

246
00:13:51,126 --> 00:13:54,636
we're buying stocks, getting full exposure
to the stock market, but we're trying

247
00:13:54,636 --> 00:13:56,166
to beat the market, like Rob mentioned.

248
00:13:57,396 --> 00:14:01,036
And I heard about
Wealthfront at some point.

249
00:14:01,086 --> 00:14:04,766
I don't know why my wife and I we
were on the websites of Wealthfront.

250
00:14:05,766 --> 00:14:12,556
And I noticed that the chief investment
officer was Bert Malkiel, who's a very

251
00:14:12,556 --> 00:14:14,336
famous, name in the world of finance.

252
00:14:14,336 --> 00:14:17,056
He wrote a book called "a
random walk down wall street".

253
00:14:17,616 --> 00:14:21,546
he's been a long time proponent
of indexing, not trying to

254
00:14:21,556 --> 00:14:22,866
beat the market, low costs.

255
00:14:23,441 --> 00:14:29,091
tax efficiency and their, VP of
research at the time, the person who

256
00:14:29,091 --> 00:14:33,281
I ended up replacing, had spent a year
at AQR also back in the early 2000s.

257
00:14:33,891 --> 00:14:37,461
So I noticed they had some very talented
researchers, very talented personnel.

258
00:14:37,461 --> 00:14:39,961
I started reading, some of the
white papers that they'd written

259
00:14:39,961 --> 00:14:42,841
in blog posts, and I was just
very impressed by the company.

260
00:14:43,601 --> 00:14:47,501
And I think a phrase that you mentioned
earlier before was one other driver for

261
00:14:47,511 --> 00:14:49,111
me, which was the average Joe thing.

262
00:14:49,881 --> 00:14:53,891
I really like the idea of having
a, serving a client base that was,

263
00:14:53,931 --> 00:14:56,971
much more directly retail at a QR.

264
00:14:56,971 --> 00:15:00,971
A lot of our clients were big
institutions like pension funds, and

265
00:15:00,991 --> 00:15:02,501
endowments, sovereign wealth funds.

266
00:15:02,931 --> 00:15:05,561
And there are individuals that are
benefiting from those investments.

267
00:15:05,811 --> 00:15:07,621
but it's just a lot more
direct at wealth front.

268
00:15:07,621 --> 00:15:10,381
We're very directly
serving the end investor.

269
00:15:11,221 --> 00:15:13,811
And what I found after joining
was it was even more rewarding.

270
00:15:13,811 --> 00:15:16,511
You'd get these emails from clients
telling us how much they love our

271
00:15:16,511 --> 00:15:18,161
product and, love the service.

272
00:15:18,161 --> 00:15:19,201
that always helps,

273
00:15:19,261 --> 00:15:20,571
that helps you sleep at night.

274
00:15:21,161 --> 00:15:25,071
so if we were to try to draw a
little bit of a landscape for robo

275
00:15:25,071 --> 00:15:26,601
advisors, where does it start?

276
00:15:26,611 --> 00:15:30,671
What would be considered like the
first robo advisor that hit the market?

277
00:15:30,811 --> 00:15:33,471
I think, Betterment
actually claims that title.

278
00:15:33,851 --> 00:15:38,481
but Betterment and Wallfront were the
two earliest ones back in, I don't

279
00:15:38,481 --> 00:15:41,581
know, 10, 12 years or so ago, I think.

280
00:15:43,476 --> 00:15:48,776
And then over time, more have
entered, I think a number have failed.

281
00:15:48,776 --> 00:15:51,386
Like Rob mentioned, it's tough
to start as a robo advisor.

282
00:15:51,386 --> 00:15:54,186
You really need to reach
scale to make it work.

283
00:15:54,786 --> 00:15:58,226
And then over the years,
there've also been a number

284
00:15:58,256 --> 00:16:00,306
of robo advisors launched by.

285
00:16:00,696 --> 00:16:02,236
larger incumbents.

286
00:16:03,156 --> 00:16:08,996
so places like Vanguard, Schwab, like
the big advisory and brokerage houses,

287
00:16:09,106 --> 00:16:13,316
have launched their own versions of robo
advisors to compliment the human financial

288
00:16:13,316 --> 00:16:16,106
advisors that they've had for decades.

289
00:16:16,656 --> 00:16:20,756
So you mentioned Vanguard and I think
at least from my experience, when I

290
00:16:20,766 --> 00:16:24,836
speak to other techies and software
engineers, the one thing that they all

291
00:16:24,836 --> 00:16:26,376
know about investing is that they should.

292
00:16:26,546 --> 00:16:29,656
Put some money on S&P
500 and keep it forever.

293
00:16:30,066 --> 00:16:32,396
And that's the smart way to do it.

294
00:16:32,396 --> 00:16:36,516
So from that point of view, if you were
to explain to a five year old software

295
00:16:36,516 --> 00:16:40,026
engineer, what's the real difference
between using something like a robot

296
00:16:40,026 --> 00:16:45,356
advisor, versus just putting some money
on S&P and hoping that the US doesn't

297
00:16:45,756 --> 00:16:48,726
stop being one of the best markets.

298
00:16:49,404 --> 00:16:52,514
I would say, first of all, even if you're
going to use Vanguard, you still have

299
00:16:52,554 --> 00:16:58,794
to make a decision about Which Vanguard
funds to buy There's S&P 500, but you

300
00:16:58,794 --> 00:17:00,394
might want to hold some bonds too.

301
00:17:00,474 --> 00:17:03,884
and how do you figure out the
mix between stocks and bonds?

302
00:17:03,884 --> 00:17:07,284
And that's probably the most important
decision an individual investor would

303
00:17:07,284 --> 00:17:11,134
make, more important than trying to
beat the market by 1%, there's no right

304
00:17:11,134 --> 00:17:14,024
or wrong answer, but what you choose
for your asset allocation is probably

305
00:17:14,034 --> 00:17:15,354
the most consequential decision.

306
00:17:15,744 --> 00:17:20,114
and we do in the book have several
chapters on this whole, idea of asset

307
00:17:20,114 --> 00:17:24,404
allocation, but, as we talked about,
there's other ways to even do better

308
00:17:24,444 --> 00:17:29,554
than just a straight Vanguard index 500
through various tax savings techniques.

309
00:17:30,131 --> 00:17:32,971
let's talk about this tax
savings techniques a little bit.

310
00:17:33,021 --> 00:17:37,671
I think you used the expression free
lunch a few times as we spoke about it.

311
00:17:38,051 --> 00:17:38,951
What's the main draw?

312
00:17:38,951 --> 00:17:43,901
What can you offer as a robo advisor
that, people might not know about?

313
00:17:43,951 --> 00:17:45,741
there's a few different chapters on these.

314
00:17:45,801 --> 00:17:48,351
I'll talk about two and then I'll
let Alex talk about one of them.

315
00:17:48,411 --> 00:17:53,071
let's say you're in the, decumulation
phase of your life where you're retired

316
00:17:53,071 --> 00:17:56,701
and you have to draw down on your
savings to pay for your expenses.

317
00:17:57,611 --> 00:18:01,901
you're faced with the decision of,
which accounts should I draw down

318
00:18:01,901 --> 00:18:03,811
from, Constantly throughout the book.

319
00:18:03,811 --> 00:18:06,601
We thought about how is what we're writing
going to be different from what you

320
00:18:06,601 --> 00:18:09,271
would get with something like ChatGPT?

321
00:18:09,866 --> 00:18:14,296
And if you ask ChatGPT, this question,
if you said, I'm an individual, I

322
00:18:14,296 --> 00:18:18,516
have $2 million in an IRAI have a
million dollars in a brokerage account.

323
00:18:18,521 --> 00:18:21,886
Some of that money has, low cost
basis, some of it has high cost basis.

324
00:18:21,891 --> 00:18:25,276
I also have a small Roth account
with a half a million dollars.

325
00:18:25,516 --> 00:18:27,196
I'm in this particular tax bracket.

326
00:18:27,246 --> 00:18:29,466
I have to take, required
minimum distributions in

327
00:18:29,466 --> 00:18:30,746
a certain number of years.

328
00:18:30,746 --> 00:18:34,306
Oh, I also have an inherited IRA
that I have to deplete in 10 years.

329
00:18:34,576 --> 00:18:37,146
how should I start withdrawing money?

330
00:18:37,336 --> 00:18:41,266
It would give you some very
generic, general advice.

331
00:18:41,276 --> 00:18:46,411
It's not really designed for giving
custom tailored, and that's what I

332
00:18:46,441 --> 00:18:49,521
think is one of the interesting things,
why I think the book was a little bit

333
00:18:49,521 --> 00:18:53,501
more interesting because you can't
just ask these questions for ChatGPT.

334
00:18:53,851 --> 00:18:55,291
and it does make a huge difference.

335
00:18:55,291 --> 00:18:59,121
Like we showed in the book that
depending on which strategy you choose

336
00:18:59,121 --> 00:19:03,641
for withdrawing your assets, it could
extend your assets by, a decade or more.

337
00:19:03,641 --> 00:19:08,491
So these are consequential decisions
that actually have a big impact.

338
00:19:08,561 --> 00:19:10,721
and it's all based on taxes.

339
00:19:11,091 --> 00:19:14,981
a second area that I mentioned in the
book, which is like the sort of free

340
00:19:14,981 --> 00:19:17,981
lunch, is this idea of asset location.

341
00:19:18,361 --> 00:19:21,681
suppose you have, a few different types
of accounts, which I just mentioned,

342
00:19:21,731 --> 00:19:27,465
maybe you have a 401k or an IRA, maybe
you have a taxable account, maybe you

343
00:19:27,465 --> 00:19:31,910
also have, a Roth account, and then you
have different assets, maybe you have,

344
00:19:32,680 --> 00:19:37,620
US stocks, international stocks, you have
different, money market, a bond account.

345
00:19:37,710 --> 00:19:41,260
how do you place those different
assets into which accounts?

346
00:19:41,660 --> 00:19:44,070
And it's a really interesting
optimization problem.

347
00:19:44,170 --> 00:19:45,700
it's not the same for everybody.

348
00:19:45,700 --> 00:19:47,580
It depends on different factors.

349
00:19:48,020 --> 00:19:49,990
And again, it's really a free lunch.

350
00:19:50,270 --> 00:19:54,920
one of the basic concepts is that
if you have money in stocks, it

351
00:19:54,920 --> 00:19:56,760
already is somewhat tax advantage.

352
00:19:56,770 --> 00:19:59,710
if you're not constantly trading the
stocks, if you are, a buy and hold

353
00:19:59,760 --> 00:20:04,490
investor, then the capital gains on
those stocks get deferred anyway, you're

354
00:20:04,490 --> 00:20:06,210
not realizing those capital gains.

355
00:20:06,450 --> 00:20:10,520
So in some ways, it might not make
sense to put those stocks in an

356
00:20:10,520 --> 00:20:14,020
IRA, which is also tax deferred.

357
00:20:14,170 --> 00:20:19,820
And even more importantly, if you
have the stocks in a regular taxable,

358
00:20:19,840 --> 00:20:23,930
Vanguard account, once you finally do
realize those capital gains, you pay the

359
00:20:23,930 --> 00:20:29,025
lower capital gains tax rates, whereas
if you had put the money in an IRA,

360
00:20:29,565 --> 00:20:33,255
when you finally take the money out,
you pay ordinary income taxes on that.

361
00:20:33,565 --> 00:20:37,795
but it's also not as simple as saying, all
right, I'm going to shift bonds into my

362
00:20:37,795 --> 00:20:40,395
IRA and stocks into my taxable account.

363
00:20:40,625 --> 00:20:41,695
It's much more complicated.

364
00:20:41,695 --> 00:20:45,225
we go through in the book, how you
would do an optimization like that.

365
00:20:46,280 --> 00:20:50,470
so just a very quick, detour for
people who are listening to this,

366
00:20:50,470 --> 00:20:56,260
who are not US-based, you mentioned
things like 401k, IRA, and Roth.

367
00:20:56,280 --> 00:20:59,350
Would you mind very quickly
explaining what these things mean?

368
00:21:00,765 --> 00:21:04,985
Yes, and in fact, we did recognize
when drafts of our book were sent to

369
00:21:04,995 --> 00:21:08,685
different, reviewers that some people
were in other countries and saying,

370
00:21:08,685 --> 00:21:10,115
why is it so focused on the US?

371
00:21:10,115 --> 00:21:13,635
So we added a couple of lines in
there to say that, even though we're

372
00:21:13,695 --> 00:21:18,790
US-centric, a lot of these same
concepts apply in other countries and

373
00:21:18,790 --> 00:21:20,600
we gave, an example or two of that.

374
00:21:21,270 --> 00:21:26,040
so in an IRA and 401k, and they're
very similar to each other, So you

375
00:21:26,040 --> 00:21:29,730
don't pay taxes when the money goes
into the IRA, but then when you

376
00:21:30,000 --> 00:21:31,910
deplete the IRA, you pay taxes.

377
00:21:32,430 --> 00:21:35,150
for a taxable account, you've
already paid taxes on the money.

378
00:21:35,420 --> 00:21:39,010
So you're using after tax money for
just a regular brokerage account.

379
00:21:39,420 --> 00:21:45,070
And at a Roth account, you also pay,
taxes up front, but then when you take

380
00:21:45,070 --> 00:21:49,610
the money out of the end, unlike an IRA,
a rough IRA, you pay no taxes at the end.

381
00:21:50,360 --> 00:21:52,830
And then there's other types of
accounts, like there's a health

382
00:21:52,880 --> 00:21:56,130
savings account where you don't pay
taxes in the beginning or at the end.

383
00:21:56,530 --> 00:21:59,660
so there's lots of different types of
accounts with different tax treatments.

384
00:22:00,308 --> 00:22:02,718
You said it was basically
an optimization problem.

385
00:22:02,718 --> 00:22:06,198
So you end up with some kind of
choice of parameters of what you

386
00:22:06,208 --> 00:22:07,698
think is more important to you.

387
00:22:08,118 --> 00:22:12,878
And then you basically can write a
program that solves that for you.

388
00:22:12,938 --> 00:22:14,198
is that roughly where it is?

389
00:22:15,268 --> 00:22:16,278
Yes, yes.

390
00:22:16,978 --> 00:22:22,138
So everybody should go to manning.com
and get the book now and go and write

391
00:22:22,218 --> 00:22:24,078
some code to optimize their money.

392
00:22:24,183 --> 00:22:25,743
Should we end the episode here then?

393
00:22:26,288 --> 00:22:27,888
it's actually a complicated problem.

394
00:22:27,938 --> 00:22:31,578
and a lot of robo advisors, do
something called taxless harvesting,

395
00:22:31,578 --> 00:22:32,758
which Alex will talk about next.

396
00:22:33,068 --> 00:22:37,178
but very few actually do asset location
because it is a very complicated

397
00:22:37,198 --> 00:22:40,068
thing to do and it is very customized.

398
00:22:40,118 --> 00:22:43,948
So surprisingly, even though
this could be a huge tax savings,

399
00:22:44,298 --> 00:22:46,248
it's not really widely done.

400
00:22:46,298 --> 00:22:49,608
and I think that's one of the exciting
things about the book is that, I think

401
00:22:49,608 --> 00:22:55,038
there will be a hockey stick, to use that
term, pattern where kind of these things,

402
00:22:55,198 --> 00:22:58,568
it might not be like a steep hockey stick,
maybe a lower slope hockey stick, but

403
00:22:58,568 --> 00:23:03,768
where these ideas start to take hold and
more and more people do them, over time.

404
00:23:04,708 --> 00:23:06,038
What's a glide path?

405
00:23:08,253 --> 00:23:11,573
so there's a whole industry of,
something called target date funds.

406
00:23:11,853 --> 00:23:13,253
Vanguard offers many of these.

407
00:23:13,273 --> 00:23:17,583
These are funds that, change
your asset allocation over time.

408
00:23:17,873 --> 00:23:22,273
So the kind of theory is that
as you get older, you should be

409
00:23:22,273 --> 00:23:24,913
holding, less stocks and more bonds.

410
00:23:25,013 --> 00:23:29,703
And we actually cover this in one
part of one chapter about how to

411
00:23:29,723 --> 00:23:31,703
actually do that optimization as well.

412
00:23:32,333 --> 00:23:36,153
and I think we didn't explicitly say
it, but between the lines, there are

413
00:23:36,153 --> 00:23:39,973
different restrictions, all these
accounts, there are only certain amounts

414
00:23:40,013 --> 00:23:43,673
and you can put in them at different
time periods and stuff like that, which

415
00:23:43,673 --> 00:23:45,563
makes it a level more complicated.

416
00:23:45,793 --> 00:23:46,373
Is that right?

417
00:23:47,223 --> 00:23:47,633
Yes.

418
00:23:47,923 --> 00:23:50,193
And there's so many complications
with all this stuff.

419
00:23:50,243 --> 00:23:54,003
we cover required minimum distribution,
so when you hold an IRA, at some point,

420
00:23:54,003 --> 00:23:58,253
you have to start taking money out of
it, and there's certain rules on that,

421
00:23:58,253 --> 00:24:02,113
and it gets even more complicated when
you talk about the rules for inherited

422
00:24:02,123 --> 00:24:07,408
IRAs, and then we mostly focused on
federal taxes, but there's also state

423
00:24:07,408 --> 00:24:09,398
taxes and every state has different rules.

424
00:24:10,068 --> 00:24:16,738
So like New York, for example, you can
take out $20k every year of an IRA without

425
00:24:16,778 --> 00:24:19,638
being subject to New York State taxes.

426
00:24:20,458 --> 00:24:23,356
So that also affects your
decision about how to start

427
00:24:23,356 --> 00:24:25,988
withdrawing money from your IRAs.

428
00:24:26,038 --> 00:24:28,658
You might want to go right up to
that limit every year, for example.

429
00:24:29,195 --> 00:24:31,615
so Monopoly all of a sudden
sounds like it's pretty

430
00:24:31,615 --> 00:24:33,185
straightforward and pretty boring.

431
00:24:33,685 --> 00:24:37,625
They should update the version
that has all of this added to it.

432
00:24:38,564 --> 00:24:38,844
I

433
00:24:39,449 --> 00:24:43,239
can definitely see the draw of
why somebody would go and pick

434
00:24:43,239 --> 00:24:45,319
up the book to learn about that.

435
00:24:45,379 --> 00:24:48,449
and we're going to get into the
technical part of the book in a sec,

436
00:24:48,489 --> 00:24:51,909
but we didn't say it explicitly either.

437
00:24:51,939 --> 00:24:55,259
Who is really your target audience?

438
00:24:56,539 --> 00:24:59,009
I think that was a really
tricky thing for Alex and I.

439
00:24:59,039 --> 00:25:04,169
So we had in mind a
widely varying audience.

440
00:25:04,199 --> 00:25:08,849
On one hand, for developers that were
interested in saving money in finance,

441
00:25:08,899 --> 00:25:12,989
or making money, it reminds me, I
was just reading the obituary of Jim

442
00:25:12,989 --> 00:25:18,839
Simons, who was like this quant guru,
and he was a Brilliant mathematician.

443
00:25:18,839 --> 00:25:21,809
And at some point in his life, he
decided, rather than working on math

444
00:25:21,809 --> 00:25:23,359
theorems, I actually want to make money.

445
00:25:23,689 --> 00:25:28,649
So part of the audience is for people
that are, coders that want to turn

446
00:25:28,649 --> 00:25:30,119
their skills into making money.

447
00:25:30,829 --> 00:25:37,117
and they have one set of skills, which is
they're really probably great at Python,

448
00:25:37,117 --> 00:25:38,894
but don't know much about the finance.

449
00:25:38,894 --> 00:25:43,809
Then we also had in mind, a segment
that maybe our, financial advisors

450
00:25:43,839 --> 00:25:48,204
that maybe, know a lot about the
finance part, but a lot of them

451
00:25:48,204 --> 00:25:51,424
actually do stuff in spreadsheets,
and they don't know any Python.

452
00:25:51,904 --> 00:25:53,364
and that's making it
even more complicated.

453
00:25:53,364 --> 00:25:55,224
The book has a little bit of math in it.

454
00:25:55,264 --> 00:25:58,794
if you were to draw a Venn diagram where
one circle is people who know Python and

455
00:25:58,794 --> 00:26:02,074
other circles, people who know finance
and other circles, people who know math,

456
00:26:02,074 --> 00:26:04,434
like the intersection of that is small.

457
00:26:04,754 --> 00:26:09,204
so we try to make it so that if you don't
know any Python and don't care to learn

458
00:26:09,204 --> 00:26:10,674
it, you could just skip the Python parts.

459
00:26:10,674 --> 00:26:14,654
If you don't know, any math, we
try to emphasize that, we include

460
00:26:14,654 --> 00:26:17,454
the math for completeness for those
people that are interested in it,

461
00:26:17,454 --> 00:26:19,234
but we could just skip that part.

462
00:26:19,234 --> 00:26:20,974
We put some of that stuff in the appendix.

463
00:26:21,539 --> 00:26:26,089
so it does have a wide audience
and in fact, I know that you

464
00:26:26,089 --> 00:26:26,899
wrote a book for Manning.

465
00:26:26,949 --> 00:26:30,319
so you're aware of this, when we
had a draft of the book, the Manning

466
00:26:30,369 --> 00:26:34,749
sends out the draft to something like
15 reviewers and Manning's staple of

467
00:26:34,749 --> 00:26:38,929
reviewers are people that are software
developers, not financial advisors,

468
00:26:38,929 --> 00:26:40,629
because that's just not their world.

469
00:26:40,629 --> 00:26:44,979
So we got, reviews from
mostly software developers.

470
00:26:45,019 --> 00:26:48,049
And in general, the reviews were great,
but obviously it's human nature to

471
00:26:48,049 --> 00:26:50,559
focus on what are some of the critiques.

472
00:26:50,599 --> 00:26:54,020
so like we had a throwaway line
in the first chapter about why we

473
00:26:54,020 --> 00:26:56,390
chose, Python over other languages.

474
00:26:56,390 --> 00:26:57,480
And they were all over that.

475
00:26:57,540 --> 00:26:59,760
we got a million comments about that.

476
00:26:59,890 --> 00:27:03,400
And then we would also get comments
like, what is a T bill or what

477
00:27:03,400 --> 00:27:04,920
does it mean to short sale?

478
00:27:05,260 --> 00:27:08,930
Alex and I had a lot of discussions about
do we want to explain these concepts?

479
00:27:08,930 --> 00:27:12,380
and usually where we landed was
that you could easily Google

480
00:27:12,450 --> 00:27:13,560
some of the financial terms.

481
00:27:13,620 --> 00:27:16,720
we didn't intend for this book to be
teaching the basics of everything.

482
00:27:16,720 --> 00:27:17,840
So we assume that people.

483
00:27:18,610 --> 00:27:20,700
have a minimum understanding of Python.

484
00:27:20,700 --> 00:27:24,440
We're not trying to teach people
Python from scratch or have a minimum

485
00:27:24,440 --> 00:27:27,040
understanding of finance and know
some of the terms and if you don't

486
00:27:27,060 --> 00:27:28,460
you just have to look those up.

487
00:27:29,735 --> 00:27:33,715
So did your target audience drift a little
bit from where you initially started

488
00:27:33,715 --> 00:27:37,435
as you were writing the book or are you
roughly where you were at the beginning?

489
00:27:38,022 --> 00:27:42,432
I think we thought a lot about drifting,
but in the end I think we ended up,

490
00:27:42,562 --> 00:27:47,002
either out of laziness, or because
this was, our philosophy, we did not

491
00:27:47,002 --> 00:27:51,152
change it that much in terms of trying
to cater to every single segment of

492
00:27:51,232 --> 00:27:54,592
the population we definitely did make
changes to try to accommodate what some

493
00:27:54,592 --> 00:27:56,492
of the reviewers were complaining about.

494
00:27:56,522 --> 00:27:59,912
I tried to make the math a little bit
simpler, but I didn't eliminate the math,

495
00:27:59,912 --> 00:28:04,432
so yeah, we definitely did a few things to
try to accommodate, some of the critiques.

496
00:28:04,657 --> 00:28:09,257
I definitely see because this really is
different to your typical manning book.

497
00:28:09,277 --> 00:28:13,537
There is just more of a completely
different domain coming in.

498
00:28:13,537 --> 00:28:17,917
So they must have been a little bit
confused at least okay, so let's move a

499
00:28:17,917 --> 00:28:21,017
little bit towards the technical bit then.

500
00:28:21,337 --> 00:28:24,107
And maybe let's talk
about some of this maths.

501
00:28:24,167 --> 00:28:28,507
I think one of the things that a lot of
software engineers have heard here and

502
00:28:28,507 --> 00:28:34,062
there, but never actually fully understood
what it's for and why it's useful, is

503
00:28:34,102 --> 00:28:36,142
things like Monte Carlo simulations.

504
00:28:36,172 --> 00:28:40,342
Can you give us like a quick, you
know, again, five year old software

505
00:28:40,342 --> 00:28:43,502
engineer listening to this wondering
what should I do with my life?

506
00:28:43,532 --> 00:28:46,272
Should I be an astronaut or
should I be a software engineer?

507
00:28:46,822 --> 00:28:47,892
Tell them about Monte Carlo.

508
00:28:48,712 --> 00:28:52,952
Monte Carlo is a hugely useful technique
for solving all sorts of problems.

509
00:28:52,952 --> 00:28:55,882
In fact, when I worked as an
engineer before I got my PhD in

510
00:28:55,882 --> 00:29:00,212
finance, I actually did Monte
Carlo simulations on radar systems.

511
00:29:00,222 --> 00:29:03,862
And then my dissertation was
related to Monte Carlo simulations.

512
00:29:03,902 --> 00:29:06,932
that was actually the title, was,
had the word Monte Carlo simulation

513
00:29:06,942 --> 00:29:08,052
in the title of the dissertation.

514
00:29:08,402 --> 00:29:10,312
so it's used in many different fields.

515
00:29:10,312 --> 00:29:13,532
and I could get into this a little
bit more later, but I think one of our

516
00:29:13,532 --> 00:29:17,412
hopes when we wrote this book is that
a lot of the techniques that we use

517
00:29:18,302 --> 00:29:21,172
are just general techniques that could
be used in a lot of different realms.

518
00:29:21,202 --> 00:29:24,992
So hopefully, the AI chapter
that I have also can be used and

519
00:29:25,002 --> 00:29:26,872
it's not specific to finance.

520
00:29:26,922 --> 00:29:28,352
it's much more general.

521
00:29:28,602 --> 00:29:33,010
so Monte Carlo is very general technique,
but it's particularly useful for solving

522
00:29:33,010 --> 00:29:39,450
certain financial planning problems,
and I show in the book that, for a

523
00:29:39,450 --> 00:29:43,410
few reasons, you might be tempted to
say, all right, stocks have an average

524
00:29:43,410 --> 00:29:49,750
return of, let's say, 10%, bonds
maybe have an average return of 4%.

525
00:29:50,170 --> 00:29:56,095
if you take a 50/50 mix of 10 percent
stocks, 4 percent bonds, that's,

526
00:29:56,095 --> 00:29:57,645
on average, 7 percent returns.

527
00:29:57,935 --> 00:30:01,615
if I'm trying to do financial planning,
can I just assume that, I, on average,

528
00:30:01,705 --> 00:30:05,805
earn 7 percent a year and just skip
the whole Monte Carlo simulation part.

529
00:30:05,805 --> 00:30:11,545
But the reasons why You can't really
do that is, first of all, you might

530
00:30:11,545 --> 00:30:15,235
want to answer questions like what's
the probability of running out of money

531
00:30:15,235 --> 00:30:18,465
in retirement, for example, and it's
not just a yes or no, it's actually

532
00:30:18,515 --> 00:30:21,705
a probability distribution of what
your final assets are going to be.

533
00:30:21,715 --> 00:30:23,205
Some of them you will run out.

534
00:30:23,265 --> 00:30:26,305
if you don't do a simulation, you won't
be able to answer probability questions.

535
00:30:26,635 --> 00:30:31,175
Also, it turns out that the order
of returns makes a difference

536
00:30:31,175 --> 00:30:32,455
when you have withdrawals.

537
00:30:32,735 --> 00:30:35,865
So if you have a good year and
then a bad year, that might

538
00:30:35,895 --> 00:30:38,225
actually be different than having
a bad year and then a good year.

539
00:30:38,225 --> 00:30:40,525
And you actually need to do
Monte Carlo simulation to

540
00:30:40,525 --> 00:30:42,405
pick up subtleties like that.

541
00:30:42,695 --> 00:30:46,545
and then, there might be certain
complications that you just can't

542
00:30:46,835 --> 00:30:50,555
figure out analytically, like with
tax rates and, things like that.

543
00:30:50,685 --> 00:30:54,215
there's no analytic closed form solutions
to figure out how much money you're going

544
00:30:54,215 --> 00:30:55,475
to have in a certain amount of time.

545
00:30:55,715 --> 00:30:56,915
it's just too complicated.

546
00:30:56,995 --> 00:31:01,115
you have to resort to Monte Carlo
simulations where you simulate, a

547
00:31:01,115 --> 00:31:05,365
whole bunch of random paths, random
stock returns, random bond returns.

548
00:31:05,555 --> 00:31:06,695
You could do anything random.

549
00:31:06,695 --> 00:31:09,615
You could have, The inflation
rate be, a random variable.

550
00:31:09,655 --> 00:31:11,465
you can even make tax brackets.

551
00:31:11,525 --> 00:31:13,755
you could randomize that and say,
there's a certain probability

552
00:31:13,755 --> 00:31:15,575
that tax brackets could change.

553
00:31:15,805 --> 00:31:19,605
So you can incorporate all those things
in a Monte Carlo simulation, generate,

554
00:31:19,605 --> 00:31:26,535
10.000 different paths, and then see what
happens, at the end of those 10.000 paths.

555
00:31:27,365 --> 00:31:28,625
So what does it actually tell you?

556
00:31:28,625 --> 00:31:33,335
if you design this and generate
10.000 different paths, and then

557
00:31:33,355 --> 00:31:37,115
you look at them in aggregate and
this is happening, most of them.

558
00:31:37,115 --> 00:31:38,955
So maybe I should worry about this.

559
00:31:39,025 --> 00:31:41,275
what output information does it give you?

560
00:31:41,675 --> 00:31:43,645
you could get a whole
probability distribution.

561
00:31:43,645 --> 00:31:45,725
You can answer questions like, I have.

562
00:31:47,035 --> 00:31:49,135
$2.000.000, what's the
probability that I'm going to

563
00:31:49,135 --> 00:31:50,725
run out of money in retirement?

564
00:31:50,965 --> 00:31:52,975
so it could give you, a probability.

565
00:31:53,335 --> 00:31:56,795
you could try different policies and
run them through Monte Carlo simulation.

566
00:31:56,795 --> 00:32:00,685
And it could tell you: on average,
policy X does better than policy Y.

567
00:32:01,995 --> 00:32:06,135
And outside of finance,
what would be some, famous

568
00:32:06,135 --> 00:32:08,485
applications that you can tell us?

569
00:32:09,070 --> 00:32:11,000
as the name implies,
gambling is always one.

570
00:32:11,220 --> 00:32:12,360
so Monte Carlo.

571
00:32:12,400 --> 00:32:15,250
but it's used all over,
in multiple fields.

572
00:32:15,485 --> 00:32:19,470
When I worked on radar systems, there's
a lot of, randomness, involved and

573
00:32:19,550 --> 00:32:21,400
there were no closed form solutions.

574
00:32:21,420 --> 00:32:24,640
the only way to, figure out, how many
targets can you track with a certain

575
00:32:24,640 --> 00:32:27,070
radar was through Monte Carlo simulation.

576
00:32:27,590 --> 00:32:29,960
we used it when I worked in
weather risk management too.

577
00:32:30,210 --> 00:32:36,570
there were underlying processes say
daily temperature and it's daily

578
00:32:36,575 --> 00:32:40,010
temperature at all these thousands
of locations across the US and the

579
00:32:40,010 --> 00:32:45,240
quantities that we were interested in
were derivations of daily temperature.

580
00:32:45,240 --> 00:32:45,960
They might be like.

581
00:32:46,865 --> 00:32:51,665
the average temperature in Chicago
over the winter, but we were also cared

582
00:32:51,665 --> 00:32:56,625
about the number of days where the
temperature dropped below zero degrees.

583
00:32:56,885 --> 00:32:58,935
some places in the Midwest
where people are farming and

584
00:32:58,935 --> 00:33:00,095
they're worried about frost.

585
00:33:01,035 --> 00:33:04,915
And those things are all related to each
other through the temperature values.

586
00:33:05,685 --> 00:33:11,035
but deriving closed form or
analytic solutions for those joint

587
00:33:11,035 --> 00:33:15,615
distributions, number of frost days
in location X and average temperature

588
00:33:15,615 --> 00:33:17,855
in location Y, is extremely difficult.

589
00:33:18,215 --> 00:33:21,575
and so running lots and lots of
simulations is a kind of a brute

590
00:33:21,575 --> 00:33:24,165
force way of doing things that
are too hard to do analytically.

591
00:33:24,715 --> 00:33:25,835
I think that makes sense.

592
00:33:25,835 --> 00:33:28,155
And that's probably going to
resonate with a lot of people.

593
00:33:28,205 --> 00:33:29,365
too much math.

594
00:33:30,095 --> 00:33:33,025
Let's just simulate some of the things
and look at it in the aggregate.

595
00:33:33,055 --> 00:33:34,315
That makes perfect sense.

596
00:33:34,795 --> 00:33:35,985
You mentioned AI.

597
00:33:36,385 --> 00:33:38,435
and I know that your book is
talking a little bit about

598
00:33:38,435 --> 00:33:40,505
the reinforcement learning.

599
00:33:40,975 --> 00:33:44,365
Can you give us a sneak peek of what
we're going to find in the book on that?

600
00:33:45,150 --> 00:33:48,730
I know this may disappoint people because
everyone just wants to talk about AI now.

601
00:33:48,730 --> 00:33:50,000
And that's the only thing
people want to talk about.

602
00:33:50,000 --> 00:33:53,700
I was just watching, Berkshire
Hathaway's annual meeting, recently,

603
00:33:53,700 --> 00:33:57,710
and people were asking the 93 year
old Warren Buffett, who doesn't even

604
00:33:57,710 --> 00:33:59,150
own a computer, what he thinks of AI.

605
00:33:59,650 --> 00:34:02,610
this is the topic that
everyone wants to talk about.

606
00:34:02,680 --> 00:34:06,500
And in fact, in the book, it was
one chapter and we didn't cover,

607
00:34:06,880 --> 00:34:08,490
generative AI in the book at all.

608
00:34:08,490 --> 00:34:12,550
And it certainly there's uses for
generative AI and financial planning.

609
00:34:12,550 --> 00:34:18,420
Like I've seen use cases where,
Generative AI can summarize meetings.

610
00:34:18,470 --> 00:34:21,200
so financial advisors love meetings
with clients and it can take the

611
00:34:21,200 --> 00:34:23,260
transcript and summarize it for people.

612
00:34:23,540 --> 00:34:24,110
it's interesting.

613
00:34:24,110 --> 00:34:31,360
I actually was listening to a webinar,
last week where, they used, a survey,

614
00:34:31,390 --> 00:34:35,930
a financial literacy survey of 38
questions that financial advisors

615
00:34:35,950 --> 00:34:40,480
sometimes give to, their clients just to
gage how financially literate they are.

616
00:34:40,870 --> 00:34:45,220
And these authors decided, let
me see how ChatGPT does with

617
00:34:45,220 --> 00:34:46,740
these set of 38 questions.

618
00:34:46,740 --> 00:34:51,590
And, ChatGPT has been famous for,
acing the bar exam and getting a

619
00:34:51,590 --> 00:34:54,300
four out of five and AP chemistry
and doing well in the MCATs.

620
00:34:54,750 --> 00:35:00,455
And it actually only got a 45% correct
on this financial literacy, which was

621
00:35:00,455 --> 00:35:04,925
about the same as how, high net worth
individuals were doing on the test that

622
00:35:04,925 --> 00:35:07,255
were not financial experts themselves.

623
00:35:07,555 --> 00:35:11,075
we didn't cover generative AI,
and I mentioned this earlier, like

624
00:35:11,085 --> 00:35:16,035
for customized non generic advice,
whether it's the best tool for that.

625
00:35:16,055 --> 00:35:21,645
there's a lot of uses for generative
AI in the, financial planning process

626
00:35:21,645 --> 00:35:24,895
and for financial advisors, but I'm
not sure it's great at certain things.

627
00:35:25,215 --> 00:35:29,995
So what I focused on was, another branch
of AI, which is, reinforcement learning.

628
00:35:30,815 --> 00:35:34,385
and one of the things I like about
the chapter is that first of all, for

629
00:35:34,675 --> 00:35:38,460
anyone who just wants to know about
what reinforcement learning is, I

630
00:35:38,480 --> 00:35:43,450
think it's a pretty good introduction
to it and a good tutorial, and I give

631
00:35:43,490 --> 00:35:46,630
several examples so you can get your
hands dirty with some really easy to

632
00:35:46,630 --> 00:35:51,170
understand examples in the finance world
of how reinforcement learning works.

633
00:35:51,610 --> 00:35:54,400
The other thing that I think is
interesting about the chapter is that

634
00:35:54,895 --> 00:35:59,605
this type of stuff is not being used
at all right now by financial advisors,

635
00:35:59,815 --> 00:36:02,335
to the extent that financial advisors
want to, it's a very competitive

636
00:36:02,335 --> 00:36:05,455
area, to the extent they want to,
differentiate themselves, I think

637
00:36:05,455 --> 00:36:07,225
it's an interesting area to do that.

638
00:36:07,705 --> 00:36:11,865
I basically go through how
reinforcement learning works.

639
00:36:11,915 --> 00:36:14,205
a lot of the chapters start with
simple examples and then they

640
00:36:14,205 --> 00:36:15,605
go to more complicated examples.

641
00:36:15,855 --> 00:36:20,794
So I start with a really simple example
of suppose that you have a million dollars

642
00:36:21,825 --> 00:36:27,015
And you have a goal that you want to have
two million dollars in ten years from now.

643
00:36:27,375 --> 00:36:32,245
What's the best way to allocate assets
between stocks and bonds to maximize

644
00:36:32,355 --> 00:36:34,115
the probability of achieving that goal?

645
00:36:34,325 --> 00:36:37,335
So the first thing I did was
actually, this type of problem

646
00:36:37,335 --> 00:36:41,465
has been solved for decades in
the economics realm using dynamic

647
00:36:41,475 --> 00:36:43,445
programming with backward recursion.

648
00:36:43,455 --> 00:36:47,410
So you would start at the terminal
state 10 years from now, and then work

649
00:36:47,420 --> 00:36:49,090
backwards and go through every node.

650
00:36:49,300 --> 00:36:52,800
you create this, state space grid
where your state might be what's

651
00:36:52,800 --> 00:36:54,550
your wealth, in any given year.

652
00:36:54,550 --> 00:36:58,930
And what year are we talking
about between years zero and 10?

653
00:36:59,240 --> 00:37:02,580
and then you have choice variables,
which is how much money do I

654
00:37:02,800 --> 00:37:04,440
allocate between stocks and bonds?

655
00:37:04,490 --> 00:37:07,340
this kind of problem was solved
using dynamic programming, and I go

656
00:37:07,340 --> 00:37:11,920
through the actual dynamic programming
solution for comparison, and then I

657
00:37:11,930 --> 00:37:14,520
solve it using reinforcement learning.

658
00:37:15,160 --> 00:37:18,120
But then I point out that the nice
thing about reinforcement learning

659
00:37:18,120 --> 00:37:22,665
and why it's such a powerful tool
is that There's several deficiencies

660
00:37:22,665 --> 00:37:24,065
of using dynamic programming.

661
00:37:24,475 --> 00:37:28,435
one of the biggest ones is the so called
curse of dimensionality, that if you

662
00:37:28,435 --> 00:37:33,925
have a lot more state variables and a
lot more decision variables, then dynamic

663
00:37:33,925 --> 00:37:37,645
programming would just be impossible
to run, it would just take too long.

664
00:37:38,025 --> 00:37:41,395
whereas you can solve those kinds of
problems using reinforcement learning.

665
00:37:42,175 --> 00:37:45,935
you may have more state variables
instead of just what your wealth is.

666
00:37:45,955 --> 00:37:47,245
You might have state variables of.

667
00:37:48,010 --> 00:37:52,160
What's your taxable wealth, and your
wealth in IRAs, you can have your

668
00:37:52,180 --> 00:37:56,240
income as a state variable, you can
have whether you've chosen, to start

669
00:37:56,240 --> 00:37:58,920
collecting Social Security yet, so you
can have all sorts of different state

670
00:37:58,920 --> 00:38:02,710
variables, and then you're forced to
do reinforcement learning if you want

671
00:38:02,710 --> 00:38:04,440
to solve that more complicated problem.

672
00:38:04,980 --> 00:38:08,820
And then in the chapter I go through
more and more complicated examples.

673
00:38:09,020 --> 00:38:11,770
And just one example that I, highlight.

674
00:38:12,790 --> 00:38:16,410
one decision that people have to make
is when to claim Social Security.

675
00:38:16,460 --> 00:38:21,090
you can tell the government anytime
between 62 and 70, when you want to start

676
00:38:21,090 --> 00:38:22,750
getting your Social Security payout.

677
00:38:22,980 --> 00:38:25,230
And the longer you wait,
the higher the payout is.

678
00:38:26,295 --> 00:38:29,355
And you can Google this,
over a dozen calculators.

679
00:38:29,395 --> 00:38:32,085
In the book I give an example
of Schwab's calculator.

680
00:38:32,245 --> 00:38:34,465
And they all pretty much
do the exact same thing.

681
00:38:34,465 --> 00:38:36,825
They do this kind of break even analysis.

682
00:38:36,915 --> 00:38:39,965
You could do this analysis
in a, two column spreadsheet.

683
00:38:39,975 --> 00:38:47,015
you basically say that, if you're going to
live to a ripe old age, and they compute

684
00:38:47,025 --> 00:38:51,635
what that cutoff is, let's say it's 84
years old, then you're better off waiting

685
00:38:51,635 --> 00:38:53,745
till you're 70 to take Social Security.

686
00:38:54,235 --> 00:38:56,495
If you're not in great health
and you think you are not going

687
00:38:56,495 --> 00:38:59,335
to live as long, either through
genetics or your own health, then

688
00:38:59,335 --> 00:39:01,195
you should try to claim at 62.

689
00:39:01,905 --> 00:39:06,265
And it's a simple calculation, but
it's a risk neutral calculation.

690
00:39:06,265 --> 00:39:08,225
It doesn't take risk into account.

691
00:39:08,225 --> 00:39:10,315
It doesn't take longevity
risk into account.

692
00:39:10,585 --> 00:39:15,145
So the techniques that I cover in that
chapter actually uses utility functions

693
00:39:15,145 --> 00:39:18,695
and takes risk into account and it can
be applied to all sorts of problems.

694
00:39:19,415 --> 00:39:25,415
some people have a defined benefit
pension plan and when you leave the

695
00:39:25,415 --> 00:39:27,755
company or retire, you're given a choice.

696
00:39:27,755 --> 00:39:30,435
Do you want it in a lump sum
or in an annuity where you get

697
00:39:30,435 --> 00:39:31,535
money for the rest of your life?

698
00:39:32,345 --> 00:39:35,805
You could solve a problem like
that, which is better using the same

699
00:39:35,825 --> 00:39:37,255
techniques in that in this chapter.

700
00:39:38,255 --> 00:39:42,065
And all of that, all the
code is in Python, right?

701
00:39:42,175 --> 00:39:42,495
Yes.

702
00:39:42,545 --> 00:39:47,085
and surprisingly the Python code
for this is ridiculously simple.

703
00:39:47,125 --> 00:39:50,315
even like neural networks, the code is
not that complicated, sometimes it's

704
00:39:50,315 --> 00:39:54,885
a little more complicated to model the
environment, but the actual reinforcement

705
00:39:54,885 --> 00:39:57,165
learning code is just a handful of lines.

706
00:39:57,165 --> 00:40:00,225
And there's a few helper functions
that are also just a handful of lines.

707
00:40:00,445 --> 00:40:04,325
So I think it's actually fairly accessible
to people to actually do this in Python.

708
00:40:05,355 --> 00:40:09,735
what's your take on why Python
seems to have completely dominated

709
00:40:09,795 --> 00:40:14,905
anything to do with finance or stats
or anything like that, it seems to

710
00:40:14,905 --> 00:40:20,005
be just obvious choice Python you
touched on it when you just threw one

711
00:40:20,005 --> 00:40:23,765
line to explain why you chose Python
and you got grief from the reviewers.

712
00:40:24,455 --> 00:40:25,655
but why do you think is that?

713
00:40:26,627 --> 00:40:28,517
it's very easy to get started with.

714
00:40:29,207 --> 00:40:32,957
the Hello World program is literally
just one line, print Hello World.

715
00:40:34,007 --> 00:40:38,637
It's popular and lots of very useful
packages have been written for it.

716
00:40:39,217 --> 00:40:44,687
So for example, in the book, we use a
convex optimization package called CVXPy.

717
00:40:45,237 --> 00:40:50,587
We use NumPy and SciPy, which
are used by so many people.

718
00:40:50,597 --> 00:40:54,257
So there's lots of support, lots of
very useful, well supported packages.

719
00:40:54,797 --> 00:40:57,087
it's not the most high performance
language in the world, but it

720
00:40:57,087 --> 00:40:58,962
can also be made pretty fast.

721
00:40:58,962 --> 00:41:02,032
if you care about that, if you're
careful with things and use some,

722
00:41:02,202 --> 00:41:06,322
again, some packages that have been
written to speed things up one other

723
00:41:06,322 --> 00:41:10,122
nice thing about Python is that it
plays pretty well with other languages.

724
00:41:10,172 --> 00:41:14,542
I've also used R a lot throughout
my career, and there's a lot of

725
00:41:14,582 --> 00:41:16,492
useful packages for R as well.

726
00:41:16,962 --> 00:41:21,062
but using R like within a system
with other programming languages

727
00:41:21,072 --> 00:41:22,372
is a lot harder than with Python.

728
00:41:23,372 --> 00:41:29,102
Okay, so I buy some of that, but to go
back to what you just said, print hello

729
00:41:29,102 --> 00:41:31,962
world is one line, so is in Fortran.

730
00:41:32,167 --> 00:41:37,157
You have to add program and end
program, but you could argue the

731
00:41:37,157 --> 00:41:41,587
same thing, and making Python fast is
typically not using Python under the

732
00:41:41,667 --> 00:41:45,177
hood, just using a C++ library that
was implemented in something else.

733
00:41:45,987 --> 00:41:50,347
So is it really just that it
looks more approachable and

734
00:41:50,347 --> 00:41:51,732
that's why everybody runs with it?

735
00:41:51,732 --> 00:41:55,187
Because I have this theory that
it's really all about brackets You

736
00:41:55,187 --> 00:41:58,887
don't use curly brackets and that's
what people typically don't see.

737
00:41:59,467 --> 00:42:00,677
And you use white space.

738
00:42:00,687 --> 00:42:02,127
So it just looks easier.

739
00:42:02,167 --> 00:42:04,807
But when you look into that,
is it that much easier to

740
00:42:04,807 --> 00:42:06,277
handle programming languages?

741
00:42:06,897 --> 00:42:09,367
We use pandas a lot in the book,
and there's a lot of things that

742
00:42:09,367 --> 00:42:10,917
are made easier when you use pandas.

743
00:42:10,937 --> 00:42:13,987
it just seems like a lot of the
tasks are just much easier done

744
00:42:14,442 --> 00:42:19,102
I know the ecosystem is fantastic, but I
suspect that the ecosystem is fantastic

745
00:42:19,192 --> 00:42:23,432
because, it got popular and because
everybody just started using that.

746
00:42:23,792 --> 00:42:30,492
So I always wonder, is it just because
of the curly brackets, but Anyway,

747
00:42:30,592 --> 00:42:34,492
so going back to your book, I was
also going to ask you why measuring

748
00:42:34,492 --> 00:42:36,732
returns is so difficult to begin with.

749
00:42:36,742 --> 00:42:41,312
Because like you said all those things
and I never really truly understood It's

750
00:42:41,412 --> 00:42:43,792
typically just some percentages, right?

751
00:42:43,792 --> 00:42:47,942
That you estimate it doesn't sound
that difficult, but then the moment

752
00:42:47,952 --> 00:42:52,032
you open a chapter about that, it's
all like variances, covariances and

753
00:42:52,032 --> 00:42:57,222
all kinds of things that either a
simple programmer like myself is

754
00:42:57,222 --> 00:42:58,792
getting a little bit intimidated by.

755
00:42:58,802 --> 00:43:03,782
So to a five year old software
engineer, can you just tell us briefly

756
00:43:03,782 --> 00:43:05,592
why it's not as simple as it looks?

757
00:43:06,937 --> 00:43:10,457
Yeah, so it's actually a rather
complicated thing, surprisingly.

758
00:43:11,077 --> 00:43:14,707
And I have, one full chapter on it,
and I cover some other issues in

759
00:43:14,707 --> 00:43:17,597
another chapter, but just returns.

760
00:43:18,247 --> 00:43:22,267
so first of all, One complication
with measuring returns in any kind

761
00:43:22,267 --> 00:43:24,797
of brokerage account, let's say
you have inflows and outflows.

762
00:43:24,827 --> 00:43:28,617
Maybe you deposit money, maybe you
withdraw money, maybe you have dividends.

763
00:43:28,657 --> 00:43:30,677
if you didn't have inflows
and outflows, it would be much

764
00:43:30,677 --> 00:43:31,927
simpler to compute returns.

765
00:43:31,927 --> 00:43:35,527
You could just look at the final value
of your assets, the beginning value,

766
00:43:35,617 --> 00:43:37,027
and that's all you need to know.

767
00:43:37,087 --> 00:43:40,937
but it turns out that, with
inflows and outflows, there's

768
00:43:41,117 --> 00:43:44,222
two methods to computing returns.

769
00:43:44,222 --> 00:43:47,972
There's, time weighted returns and
money or dollar weighted returns.

770
00:43:48,222 --> 00:43:53,522
So imagine as a simple example that
you, started out with some amount

771
00:43:53,522 --> 00:43:58,252
of money, let's say a million
dollars and had great returns.

772
00:43:58,252 --> 00:44:00,152
you earn 20 percent that year.

773
00:44:00,512 --> 00:44:02,492
And then let's say you
add it to your account.

774
00:44:02,492 --> 00:44:06,602
Let's say you deposited another million
dollars and then you have a bad year.

775
00:44:06,602 --> 00:44:07,722
Maybe you were down...

776
00:44:08,862 --> 00:44:11,662
15% what would you say your return is?

777
00:44:11,662 --> 00:44:15,632
on one hand you had a 20 percent
return and then a 15 percent return.

778
00:44:15,632 --> 00:44:17,512
So your returns seem to be positive.

779
00:44:17,702 --> 00:44:20,832
On the other hand, your ending value
is lower than your starting value.

780
00:44:20,832 --> 00:44:24,162
It could be, than the amount of
money you put in because when

781
00:44:24,162 --> 00:44:27,022
you had the bad returns, it was
on a larger amount of money.

782
00:44:27,282 --> 00:44:30,452
So that basically is the difference
between, In the first case,

783
00:44:30,452 --> 00:44:31,582
it's time weighted returns.

784
00:44:31,802 --> 00:44:34,152
In the second case, it's
dollar weighted returns.

785
00:44:34,552 --> 00:44:37,282
people might not notice this, but when
you look at your Vanguard statement,

786
00:44:37,502 --> 00:44:39,712
it will tell you your returns,
and there's, a little footnote.

787
00:44:40,032 --> 00:44:42,812
And if you go into the footnote,
it will say, in Vanguard's case,

788
00:44:43,172 --> 00:44:45,157
they use Dollar weighted returns.

789
00:44:45,157 --> 00:44:47,777
In other cases, they use
time weighted returns.

790
00:44:47,797 --> 00:44:50,117
and people probably don't
even look at the footnotes.

791
00:44:50,197 --> 00:44:53,257
You don't even understand what it is,
but it actually makes a difference.

792
00:44:53,727 --> 00:44:58,797
The second issue is if you're evaluating
investments, and that's a huge thing

793
00:44:59,157 --> 00:45:03,897
in my other sort of career of hedge
fund strategies, that's a huge issue.

794
00:45:03,897 --> 00:45:07,392
But even if you're looking at
robo advisors and some robo

795
00:45:07,392 --> 00:45:10,742
advisors says, we perform better.

796
00:45:11,102 --> 00:45:14,332
you have to evaluate performance and
you can't simply look at returns.

797
00:45:14,332 --> 00:45:16,692
You have to look at risk adjusted returns.

798
00:45:17,072 --> 00:45:23,127
it may be the case that, one
advisor has, on average, 10 percent

799
00:45:23,157 --> 00:45:25,387
returns, but wildly volatile.

800
00:45:25,577 --> 00:45:29,697
And another advisor maybe only has 6
percent returns, but they're basically

801
00:45:29,697 --> 00:45:31,927
like a bond, and bonds are only paying 5%.

802
00:45:31,927 --> 00:45:34,487
So they're outperforming bonds
with actually no extra risk.

803
00:45:34,907 --> 00:45:38,917
So you have to factor risk into
account when you're evaluating returns.

804
00:45:38,937 --> 00:45:41,917
And there's several
methods for doing that.

805
00:45:41,917 --> 00:45:46,187
in a chapter, I cover, sharp ratio, which
is one of the more common measures of

806
00:45:46,187 --> 00:45:51,597
risk adjusted returns, and alpha, which
is, made it to the popular press as well.

807
00:45:51,597 --> 00:45:55,627
And alpha is also another way of
evaluating risk adjusted returns.

808
00:45:57,192 --> 00:46:00,322
In fact, when Google announced that
they changed their name from Google to

809
00:46:00,322 --> 00:46:03,502
Alphabet, in their press release, they
said, part of the reason is because

810
00:46:03,502 --> 00:46:07,282
we're trying to catalog the entire,
alphabet of the world, but also we think

811
00:46:07,422 --> 00:46:10,462
of, of our company taking Alpha bets.

812
00:46:10,732 --> 00:46:13,812
so that's another reason why
they call their company Alphabet.

813
00:46:15,232 --> 00:46:18,472
So all the chasing alpha that
all comes from that theory.

814
00:46:19,097 --> 00:46:19,517
yes.

815
00:46:20,752 --> 00:46:24,322
And I give an example in the book
where I actually apply this principle.

816
00:46:24,412 --> 00:46:29,272
one very popular, area of
investing now is ESG investing,

817
00:46:29,272 --> 00:46:30,822
environmental, social, and governance.

818
00:46:30,832 --> 00:46:36,292
So people who feel like they want to
not just invest in the S&P 500, but they

819
00:46:36,292 --> 00:46:40,342
want to invest in companies that do good,
either in terms of the environment, in

820
00:46:40,342 --> 00:46:42,592
terms of how they pay their workers, etc.

821
00:46:43,252 --> 00:46:46,322
And there's several funds
that cater to those investors.

822
00:46:46,622 --> 00:46:51,772
So I actually go through a case study
of trying to evaluate the returns of an

823
00:46:51,782 --> 00:46:53,682
ESG fund that's been around for a while.

824
00:46:55,027 --> 00:47:01,647
One good way to intimidate your reader
is to name a chapter after a model or a

825
00:47:01,647 --> 00:47:04,447
theorem that has two names, hyphenated.

826
00:47:05,177 --> 00:47:07,777
In your case, it's the
Black-Litterman model.

827
00:47:08,817 --> 00:47:13,377
a challenge for you to explain
that in 30 seconds or less,

828
00:47:13,777 --> 00:47:14,367
Okay.

829
00:47:14,797 --> 00:47:16,977
and then send people
to the actual chapter.

830
00:47:17,297 --> 00:47:21,877
So the goal of the Black-Litterman
model was to accomplish two things.

831
00:47:21,897 --> 00:47:28,297
One is to overcome the sensitivity that's
inherent to portfolio optimization.

832
00:47:28,917 --> 00:47:32,487
if you're using portfolio optimization
and you tweak the inputs just

833
00:47:32,497 --> 00:47:35,957
slightly, you can end up with two
completely different portfolios.

834
00:47:36,007 --> 00:47:39,327
And the other goal was
to be able to allow.

835
00:47:40,137 --> 00:47:47,967
Investors to specify their views about how
they think certain assets will perform,

836
00:47:48,387 --> 00:47:53,617
either in an absolute sense or relative
to one another, the model supports both.

837
00:47:53,997 --> 00:47:57,017
the way the model works is by starting
with kind of a reference portfolio

838
00:47:57,197 --> 00:48:02,607
from which the investors derive
equilibrium returns, and then they

839
00:48:02,647 --> 00:48:07,497
add in and blend in their views about
returns to those equilibrium returns.

840
00:48:07,847 --> 00:48:10,967
The equilibrium returns keep the
investors optimized portfolio anchored

841
00:48:11,547 --> 00:48:16,177
to the reference portfolio, but then
the views, the second part of the

842
00:48:16,177 --> 00:48:20,757
model, allowing the investor to specify,
their views on the, future returns

843
00:48:20,757 --> 00:48:25,677
of the assets that they're investing
in, allow them, some deviation from

844
00:48:25,707 --> 00:48:27,787
just The standard reference portfolio,

845
00:48:28,787 --> 00:48:31,727
Okay, I'm not gonna lie, you
did lose me a little bit there.

846
00:48:31,797 --> 00:48:34,467
Is there a way to explain
it in even simpler terms?

847
00:48:35,437 --> 00:48:44,727
let's see a way to allow investors
to build portfolios that incorporate

848
00:48:44,737 --> 00:48:47,847
their views about the expected
returns of the assets they're

849
00:48:47,857 --> 00:48:53,867
investing in while combating some
of the Input sensitivity that's just

850
00:48:53,867 --> 00:48:55,737
inherent to portfolio optimization.

851
00:48:56,237 --> 00:48:56,867
Got it.

852
00:48:57,007 --> 00:49:02,987
Another thing I was gonna ask you about
is, tax loss harvesting, because it makes

853
00:49:03,027 --> 00:49:07,517
sense to me, it sounds fruit harvesting,
and that's a good thing, but at the

854
00:49:07,517 --> 00:49:10,657
same time, why would you harvest loss?

855
00:49:11,417 --> 00:49:12,617
Why is it a good idea?

856
00:49:12,627 --> 00:49:13,687
What does it give you?

857
00:49:13,827 --> 00:49:18,417
So the phrase that you always hear
in finances is buy low, sell high.

858
00:49:18,427 --> 00:49:21,967
And that's typically what you
want to do, but with tax loss

859
00:49:21,967 --> 00:49:25,997
harvesting, what you're actually
doing is selling low intentionally.

860
00:49:26,887 --> 00:49:28,387
but that can actually be a good thing.

861
00:49:29,257 --> 00:49:32,307
I think maybe the best way to
explain it is through an example.

862
00:49:33,197 --> 00:49:38,777
Oh, you mentioned the S&P 500 before
that's, an index of large cap US stocks.

863
00:49:39,892 --> 00:49:42,412
And there's another index out
there called the Russell 1000.

864
00:49:43,012 --> 00:49:46,832
it's 1000, stocks instead of
500, but those two things are

865
00:49:46,952 --> 00:49:48,302
super, super highly correlated.

866
00:49:49,142 --> 00:49:53,622
If the S&P is down 2 percent one day,
you can bet that the Russell 1000

867
00:49:53,622 --> 00:49:55,472
is going to be down also, around 2%.

868
00:49:57,112 --> 00:50:02,372
So the idea of tax loss harvesting is
opportunistically selling assets at

869
00:50:02,372 --> 00:50:05,672
a loss to reduce your current taxes.

870
00:50:05,722 --> 00:50:11,332
And the way that most robo advisors
do it is by using pairs of funds or

871
00:50:11,332 --> 00:50:16,942
pairs of ETFs that follow different
indices, but indices that are very

872
00:50:16,942 --> 00:50:22,307
highly correlated so that when you sell
one and buy the other, your economic

873
00:50:22,307 --> 00:50:24,387
exposure really doesn't change very much.

874
00:50:24,737 --> 00:50:28,657
So going back to our example, if you
bought S&P 500 at a hundred dollars,

875
00:50:29,657 --> 00:50:33,297
and then a few months later, it was
down to 90, you sell $90 of that

876
00:50:33,297 --> 00:50:36,217
fund, you've realized $10 in losses.

877
00:50:37,077 --> 00:50:40,007
Then you buy $90 worth of
their Russell 1000 fund.

878
00:50:41,017 --> 00:50:43,757
Your economic exposure
is basically unchanged.

879
00:50:43,797 --> 00:50:47,217
You're still invested in $90
worth of, US large cap stocks.

880
00:50:48,207 --> 00:50:50,387
Oh, but you've realized $10 in losses.

881
00:50:50,397 --> 00:50:54,877
So you can multiply that by your tax
rate that applies to that loss, which

882
00:50:54,877 --> 00:50:59,567
might be like 40 percent because it's
a short term loss, that means that if

883
00:50:59,567 --> 00:51:04,527
you have capital gains elsewhere in your
portfolio, you can use those losses to

884
00:51:04,527 --> 00:51:08,537
offset those gains and then decrease
your tax bill in the current year.

885
00:51:09,927 --> 00:51:11,647
So you save money on your current taxes.

886
00:51:12,587 --> 00:51:17,457
And that's the very short, simple
goal of tax loss harvesting.

887
00:51:18,257 --> 00:51:22,187
and that sounds like a, a free lunch,
but it actually isn't because, if

888
00:51:22,187 --> 00:51:27,587
you fast forward 10 years, what
happens, let's say the price of

889
00:51:28,287 --> 00:51:30,867
.,
either of those index
funds has gone up to $200.

890
00:51:32,137 --> 00:51:35,897
If you had just stuck with your original
investment, you would have had a gain

891
00:51:35,907 --> 00:51:40,527
of $100 right over that, that 10 years,
because you've harvested the loss.

892
00:51:41,272 --> 00:51:43,882
You bought the second ETF for $90.

893
00:51:43,882 --> 00:51:47,062
So now 10 years later, you
have a gain of $110 instead.

894
00:51:47,272 --> 00:51:49,212
So you've decreased your current

895
00:51:49,472 --> 00:51:52,932
taxes, but that comes at
the cost of increasing your

896
00:51:52,932 --> 00:51:55,032
taxes way off in the future.

897
00:51:56,547 --> 00:51:58,972
but that's still a good thing
because the other phrase that you

898
00:51:58,972 --> 00:52:02,762
hear in finances, A dollar today is
worth, more than a dollar tomorrow.

899
00:52:03,212 --> 00:52:04,562
It's just kinda the opposite with taxes.

900
00:52:04,862 --> 00:52:09,912
Paying $110 in taxes 10 years in the
future it's worth paying the extra

901
00:52:09,912 --> 00:52:14,022
taxes 10 years later to save money
on taxes now because you can reinvest

902
00:52:14,027 --> 00:52:18,082
that money now and then in the future,
you'll have more money than you would

903
00:52:18,082 --> 00:52:22,682
have if you had not done the tax loss
harvesting followed by the reinvestment.

904
00:52:23,682 --> 00:52:29,552
Creating kind of almost artificial
loss that, lets you defer some of the

905
00:52:29,552 --> 00:52:33,502
taxes that you, we would have to pay
otherwise, and you can invest them in

906
00:52:33,502 --> 00:52:35,572
the meantime and, be more efficient.

907
00:52:36,392 --> 00:52:37,622
That sounds pretty creative.

908
00:52:37,632 --> 00:52:41,482
and that's another thing that
you can learn from the book.

909
00:52:41,602 --> 00:52:45,202
The book is called "build
a robo advisor with Python.

910
00:52:45,672 --> 00:52:50,102
From scratch" and it's available
at manning.com a quick question.

911
00:52:50,502 --> 00:52:54,122
I've been wondering about it for
probably at least last 45 minutes.

912
00:52:54,152 --> 00:52:58,752
Is it true that like that tech people,
I'm going to just bundle everybody,

913
00:52:58,752 --> 00:53:00,882
software engineers, all the programmers.

914
00:53:01,882 --> 00:53:05,622
There's definitely this trend where
this people like optimizing things

915
00:53:05,622 --> 00:53:09,452
in their lives, whether it's diet,
maybe working out and stuff like that.

916
00:53:09,502 --> 00:53:14,992
But from where you sit, are they any
better at optimizing their savings and

917
00:53:15,002 --> 00:53:19,332
their personal finance than the average
Joe who doesn't happen to be a programmer?

918
00:53:19,882 --> 00:53:24,872
to optimize this stuff is very hard to
do on your own, which is why it's good

919
00:53:24,882 --> 00:53:27,462
to have, this done in an automated way.

920
00:53:27,542 --> 00:53:29,922
We didn't cover this, but in the
first chapter of the book, we talk

921
00:53:29,932 --> 00:53:36,832
about how, individual investors
suffer from various behavioral biases.

922
00:53:37,132 --> 00:53:39,802
and I think tech investors
probably suffer from the same

923
00:53:40,032 --> 00:53:42,092
behavioral biases as any investors.

924
00:53:42,092 --> 00:53:45,202
And some of those examples, and
there's been studies on this.

925
00:53:45,202 --> 00:53:49,197
First of all, Individual
investors tend to overtrade.

926
00:53:49,237 --> 00:53:52,167
it's generally not a great
idea to trade too much.

927
00:53:52,167 --> 00:53:54,587
You pay taxes, you pay transaction costs.

928
00:53:54,587 --> 00:53:57,807
Even if there's no commissions, there's
bid ask spreads and things like that.

929
00:53:58,057 --> 00:54:01,637
so overtrading is typically a bad
thing and individual investors

930
00:54:01,927 --> 00:54:03,107
typically do too much of it.

931
00:54:04,002 --> 00:54:07,442
there's also another, behavioral
bias that individual investors

932
00:54:07,482 --> 00:54:09,092
suffer from is herding.

933
00:54:09,092 --> 00:54:13,932
So it's typical that when things
keep going up and up, people want

934
00:54:13,932 --> 00:54:17,472
to buy it and when things go down,
people panic and want to sell.

935
00:54:17,642 --> 00:54:22,382
And that's not necessarily, empirically
the best strategy to follow.

936
00:54:22,692 --> 00:54:26,812
So again, a robo advisor can, eliminate
some of these behavioral biases.

937
00:54:26,812 --> 00:54:27,522
if some things.

938
00:54:28,442 --> 00:54:31,492
gone down, they might
buy a little bit of it.

939
00:54:31,842 --> 00:54:35,462
They're not going to sell it as a panic
sell, and they might actually buy a little

940
00:54:35,472 --> 00:54:41,142
bit, to rebalance, which empirically works
out, better than, this herding mentality.

941
00:54:41,492 --> 00:54:44,742
and there's other behavioral biases,
individual investors, and again, I don't

942
00:54:44,742 --> 00:54:46,152
think tech investors are any different.

943
00:54:46,382 --> 00:54:48,662
They don't like to, realize losses.

944
00:54:48,682 --> 00:54:53,002
if they've had a stock that's gone
down, they, it's like admitting

945
00:54:53,002 --> 00:54:54,842
defeat by selling it at a loss.

946
00:54:54,912 --> 00:54:55,892
They just don't like doing it.

947
00:54:56,042 --> 00:54:57,762
There's a lot of studies on this.

948
00:54:57,832 --> 00:55:02,132
and, but for Robo Advisor, they
actually say, Oh, we actually

949
00:55:02,132 --> 00:55:03,442
do want to sell it at a loss.

950
00:55:03,712 --> 00:55:05,412
We're going to buy something back similar.

951
00:55:05,442 --> 00:55:08,192
but we don't have any
problem realizing this loss.

952
00:55:08,202 --> 00:55:12,572
so I think that I'm not sure that
tech investors are any better

953
00:55:12,612 --> 00:55:13,902
doing this than anyone else.

954
00:55:13,912 --> 00:55:16,752
And we know that general investors
are not very good at this.

955
00:55:17,272 --> 00:55:22,692
I think tech people might be more
willing to embrace tools that

956
00:55:23,022 --> 00:55:25,912
help them avoid those sorts of
biases than other people might.

957
00:55:26,912 --> 00:55:31,452
So there might be overly
represented in the client base of

958
00:55:31,492 --> 00:55:33,812
robo advisors at the very least.

959
00:55:34,002 --> 00:55:34,652
that's what we see.

960
00:55:36,312 --> 00:55:36,782
All right, great.

961
00:55:36,792 --> 00:55:38,752
where do you see all of this going?

962
00:55:38,752 --> 00:55:43,242
So to me, it sounds like there are certain
things that can be automated and you

963
00:55:43,242 --> 00:55:45,772
can be more efficient with your taxes.

964
00:55:46,352 --> 00:55:48,112
You can harvest some of those losses.

965
00:55:48,112 --> 00:55:53,082
You can do all kinds of things to,
to optimize the way, the timing and

966
00:55:53,122 --> 00:55:55,162
the things that we just discussed.

967
00:55:55,772 --> 00:55:58,602
what's the next step for robo advisors?

968
00:55:58,812 --> 00:56:01,172
Is that the final destination?

969
00:56:01,232 --> 00:56:05,672
Is that what they're going to stay,
doing or is there like the next step

970
00:56:05,672 --> 00:56:09,352
where they can provide even more value
for like the next generation of it?

971
00:56:09,447 --> 00:56:11,787
I have one small comment and
then I'll let Alex do the rest.

972
00:56:11,847 --> 00:56:16,737
so when I had a brief, stint at a
robo advisor, we would do these focus

973
00:56:16,737 --> 00:56:20,957
groups where we would, have people
come in, we pay them a hundred dollars

974
00:56:20,957 --> 00:56:24,857
for a few hours and we would show them
our website and have them comment.

975
00:56:24,877 --> 00:56:28,847
And it was fascinating for me to
actually hear what like individual

976
00:56:28,847 --> 00:56:30,287
investors, what they care about.

977
00:56:30,647 --> 00:56:32,787
And a lot of them actually care about.

978
00:56:33,177 --> 00:56:33,957
beating the market.

979
00:56:33,957 --> 00:56:36,207
And we would try to say
no, that's not our thing,

980
00:56:37,267 --> 00:56:39,907
Rightly or wrongly, I think
that's what people want.

981
00:56:39,917 --> 00:56:44,397
they already go maybe into
what's called smart beta ETFs.

982
00:56:44,407 --> 00:56:48,827
These are ETFs that, claim that
they, based on historical anomalies,

983
00:56:48,957 --> 00:56:51,417
try to outperform the market.

984
00:56:51,477 --> 00:56:54,677
So I think that could be an area.

985
00:56:55,552 --> 00:57:03,502
I think, besides just investments,
perhaps expansion into more financial

986
00:57:03,832 --> 00:57:09,562
products that can benefit from technology,
either to lower the cost or the

987
00:57:09,562 --> 00:57:11,702
efficiency or the headache for people.

988
00:57:12,962 --> 00:57:16,382
would be one way that I could
see robo advisors expanding.

989
00:57:17,557 --> 00:57:20,627
probably another thing to consider
that we've touched on during this

990
00:57:20,627 --> 00:57:24,377
call is that there's some things that
robo advisors just will never do.

991
00:57:24,637 --> 00:57:30,477
There are certain situations that
really need hands on, advice, like

992
00:57:30,477 --> 00:57:34,647
a state planning, things like that,
where it's very specific to the person.

993
00:57:35,067 --> 00:57:37,397
It can be lots of rules,
regulations, very specific in

994
00:57:37,397 --> 00:57:41,127
that you're hard for a program.

995
00:57:41,507 --> 00:57:47,297
To consider all of the possible variations
and, specific situations of people.

996
00:57:47,897 --> 00:57:51,327
so I think there's always going
to be a place for human, very

997
00:57:51,327 --> 00:57:52,757
hands-on, tailored advice.

998
00:57:53,647 --> 00:57:59,157
but, hopefully robo advisors
gain a wider share in what can

999
00:57:59,157 --> 00:58:00,907
be automated, a lot more easily.

1000
00:58:01,907 --> 00:58:07,572
So would you say that In, the famous
80/20 rule, you get 80 percent of the

1001
00:58:07,582 --> 00:58:12,792
benefits that you can get from, just being
a bit more aware of what you do with your

1002
00:58:12,792 --> 00:58:18,172
money with 20 percent of effort or is
it maybe even better in 90/10, just by

1003
00:58:18,182 --> 00:58:20,102
using something like that to begin with.

1004
00:58:20,112 --> 00:58:22,392
Yeah, I don't know about the
exact split, but I'd say that

1005
00:58:22,392 --> 00:58:23,822
rule definitely applies, right?

1006
00:58:23,962 --> 00:58:29,282
Most of my life I'm going to be
saving and investing, and then there's

1007
00:58:29,332 --> 00:58:33,352
only going to be one point in my
life where I have to, plan for what

1008
00:58:33,362 --> 00:58:35,452
happens after I die, for example.

1009
00:58:36,402 --> 00:58:38,763
No robo advisor can help me with
the whole thing up until the

1010
00:58:38,763 --> 00:58:39,672
point where I have to decide

1011
00:58:40,172 --> 00:58:45,892
Rob, a question to you, probably more
given, you're educating people in finance.

1012
00:58:46,652 --> 00:58:51,022
Would you say that the trend of people
getting more aware of what happens to

1013
00:58:51,022 --> 00:58:55,632
their money and being, more financially
savvy, and invest in different

1014
00:58:55,642 --> 00:58:59,582
things is the financial literacy.

1015
00:58:59,582 --> 00:59:03,412
I guess it's the word really to use
here, getting more widespread or

1016
00:59:03,412 --> 00:59:07,522
is it still staying something that
the biggest chunk of the population

1017
00:59:07,532 --> 00:59:08,932
will probably never bother with.

1018
00:59:09,810 --> 00:59:13,820
I've been teaching at NYU for, 16
years or so, and I've definitely

1019
00:59:13,820 --> 00:59:17,130
noticed differences over the years I
don't know if I could speak to like

1020
00:59:17,150 --> 00:59:20,530
whether the general population is
getting more financially savvy or not.

1021
00:59:20,610 --> 00:59:25,400
my guess would be no, but there's
definitely been changes in what students

1022
00:59:25,410 --> 00:59:27,200
bring to the table over the years.

1023
00:59:27,250 --> 00:59:30,620
When I started out in finance,
it was really hard to like,

1024
00:59:31,250 --> 00:59:34,740
know the institutional
details of how markets worked.

1025
00:59:36,125 --> 00:59:39,725
It was like almost there were like a few
people that kind of understood how like

1026
00:59:39,725 --> 00:59:43,045
the specialist system in the New York
Stock Exchange worked and, the professors

1027
00:59:43,085 --> 00:59:46,375
that knew that you would look up to them,
like, how did they figure all this out,

1028
00:59:46,375 --> 00:59:49,515
like how they know both the academics
and like the institutional knowledge.

1029
00:59:49,545 --> 00:59:52,805
And now I think the institutional
knowledge is just much more

1030
00:59:52,815 --> 00:59:56,055
widely known, through, the
internet and things like that.

1031
00:59:56,130 --> 01:00:01,160
So is the right word to say commoditizing
that kind of knowledge, or is it

1032
01:00:01,170 --> 01:00:02,750
just the internet doing its thing?

1033
01:00:02,760 --> 01:00:06,050
People talk and secrets spread out,

1034
01:00:06,505 --> 01:00:07,255
that's also true.

1035
01:00:07,325 --> 01:00:12,155
when I started, it was much harder to even
know, like what some hedge funds do now

1036
01:00:12,155 --> 01:00:16,435
there's still a lot of secrecy in hedge
funds, like nobody knows how Jim Simons,

1037
01:00:16,575 --> 01:00:20,925
earns 66 percent returns every year, like
he doesn't go out and, tell people like

1038
01:00:20,925 --> 01:00:25,565
exactly the secret sauce of what he does,
but I think there's a lot more knowledge

1039
01:00:25,805 --> 01:00:31,700
about what hedge funds do and how they
make money, for a variety of reasons.

1040
01:00:31,790 --> 01:00:35,630
yeah, there was, something called
the quant crash in 2007 and

1041
01:00:35,640 --> 01:00:38,460
all the hedge funds got killed,
over this three day period.

1042
01:00:38,670 --> 01:00:42,010
And then people reverse engineered it and
figured out, Oh, all these hedge funds

1043
01:00:42,010 --> 01:00:43,950
are doing the, very similar strategies.

1044
01:00:43,950 --> 01:00:47,450
I think thas che other thing
with in the hedge fund world is

1045
01:00:47,560 --> 01:00:50,750
it's getting much harder to find,
great opportunities to make money.

1046
01:00:50,800 --> 01:00:52,050
I guess this is always the case.

1047
01:00:52,050 --> 01:00:55,850
Like when I started out, in the hedge
fund world, I traded many different

1048
01:00:55,850 --> 01:00:59,990
strategies, but one strategy I traded
was an index rebalancing strategy.

1049
01:01:00,000 --> 01:01:04,290
So all the indices like the S&P 500
would periodically rebalance and

1050
01:01:04,300 --> 01:01:09,000
you could predict what the S&P 500
would have to do, in one week's time.

1051
01:01:09,000 --> 01:01:15,320
And there's a whole cottage industry of
hedge funds that would try to They'd say,

1052
01:01:15,320 --> 01:01:17,710
Oh, the hedge funds need to buy the stock.

1053
01:01:17,750 --> 01:01:19,390
We're going to buy this
in advance of them.

1054
01:01:19,840 --> 01:01:22,000
that was like some knowledge
that a lot of people know

1055
01:01:22,000 --> 01:01:23,780
about now it's just widespread.

1056
01:01:23,780 --> 01:01:26,240
they talk about it on
CNBC, it's everywhere.

1057
01:01:26,240 --> 01:01:31,580
So a lot of these types of things
just become much more well known.

1058
01:01:33,305 --> 01:01:33,795
got it.

1059
01:01:35,215 --> 01:01:35,955
Your book.

1060
01:01:36,065 --> 01:01:40,225
I see that the publication
is estimated for June, 2024.

1061
01:01:40,225 --> 01:01:41,765
So I'm guessing it's all finished.

1062
01:01:41,765 --> 01:01:43,295
You're doing the final touches.

1063
01:01:43,735 --> 01:01:45,605
Do you know when it's
supposed to hit Amazon?

1064
01:01:47,000 --> 01:01:49,940
I think the end of June,
24th or 25th is the date.

1065
01:01:50,353 --> 01:01:53,403
we had the copy editor go through
the whole book, which she did a

1066
01:01:53,403 --> 01:01:56,553
very good job, and now we have the
typesetter going through the book.

1067
01:01:57,803 --> 01:01:58,183
awesome.

1068
01:01:58,193 --> 01:02:01,843
So sometime in July if you're on
LinkedIn, you're going to see plenty

1069
01:02:01,843 --> 01:02:07,203
of selfies with the book I was
like, oh, it's real Looks like this

1070
01:02:07,623 --> 01:02:08,633
Congratulations.

1071
01:02:08,643 --> 01:02:08,923
So

1072
01:02:08,943 --> 01:02:11,333
to write good reviews on
it and do all that stuff.

1073
01:02:11,333 --> 01:02:14,233
Yeah.

1074
01:02:14,893 --> 01:02:17,943
you can have a lot of explaining
of how to write a review on Amazon.

1075
01:02:17,943 --> 01:02:19,943
So I'm hoping you ready for that

1076
01:02:22,313 --> 01:02:23,583
What's next after that?

1077
01:02:23,593 --> 01:02:25,763
Do you have an eye for the next adventure?

1078
01:02:26,533 --> 01:02:28,333
Are you teaming up for another book?

1079
01:02:28,743 --> 01:02:29,863
Or taking a breather?

1080
01:02:31,168 --> 01:02:32,368
I would consider another book.

1081
01:02:32,368 --> 01:02:33,888
I haven't figured out the topic yet.

1082
01:02:34,268 --> 01:02:35,758
It's like that Steven Wright joke.

1083
01:02:35,758 --> 01:02:37,928
It's an old joke, but I've
written the page numbers now.

1084
01:02:37,928 --> 01:02:39,108
I just have to fill in the rest.

1085
01:02:39,428 --> 01:02:43,598
but I'm also, I got the domain
name pynancial.com and I hope

1086
01:02:43,598 --> 01:02:47,818
to do some kind of blog where I
periodically post, Interesting.

1087
01:02:48,698 --> 01:02:52,588
articles on the intersection
of finance and Python.

1088
01:02:52,658 --> 01:02:55,048
I'm excited about, posting to there, too.

1089
01:02:56,565 --> 01:02:57,515
Pynantial.

1090
01:02:58,298 --> 01:03:00,078
Yeah, I like financial, but PYNANCIAL.

1091
01:03:00,818 --> 01:03:04,188
PYNANCE was PYNANCIAL
was the next best thing.

1092
01:03:04,198 --> 01:03:05,488
we'll be on the lookout for that.

1093
01:03:05,508 --> 01:03:06,658
What about you, Alex?

1094
01:03:06,948 --> 01:03:10,458
let's see, this book was way more work
than I thought it was going to be.

1095
01:03:10,598 --> 01:03:12,978
maybe I'll do another one,
but it'll probably be a while.

1096
01:03:13,468 --> 01:03:16,378
one thing I am interested in
though, is, teaching along the

1097
01:03:16,378 --> 01:03:17,738
lines of what Rob is doing.

1098
01:03:17,738 --> 01:03:20,048
So that might be my next adventure.

1099
01:03:20,828 --> 01:03:22,388
Yeah, I had the same thing.

1100
01:03:22,388 --> 01:03:26,658
So I was writing my first book with
manning over the pandemic period.

1101
01:03:26,708 --> 01:03:31,268
And, I would just wake up at five
and I have a few hours before I

1102
01:03:31,268 --> 01:03:33,308
started working and working from home.

1103
01:03:33,308 --> 01:03:37,698
So I basically use my commute time
to write the bulk of it, really.

1104
01:03:38,328 --> 01:03:42,578
And then, one day it was finished,
and I didn't have to wake up at 5,

1105
01:03:42,578 --> 01:03:44,198
and I felt a little weird about it.

1106
01:03:44,228 --> 01:03:47,438
And I'm missing that, so I'm
guessing that's gonna kick in

1107
01:03:48,078 --> 01:03:49,528
soon enough for the two of you.

1108
01:03:49,863 --> 01:03:50,573
We'll see when it kicks in.

1109
01:03:51,033 --> 01:03:51,723
All right, guys.

1110
01:03:52,373 --> 01:03:54,643
It's been an absolute
pleasure to talk to you.

1111
01:03:54,818 --> 01:03:59,898
I think, you've achieved in educating,
at least at the very minimum, myself

1112
01:03:59,898 --> 01:04:04,008
about a lot of these things and
hopefully some of our listeners as well.

1113
01:04:04,368 --> 01:04:08,328
once again, the book's available at
manning.com at the moment, it's still

1114
01:04:08,328 --> 01:04:11,868
in the early access, program, but all
the chapters are available right now.

1115
01:04:11,928 --> 01:04:12,978
You can go and have a browse.

1116
01:04:13,428 --> 01:04:17,108
And, from what it sounds like in a
couple of months, you too will be able

1117
01:04:17,108 --> 01:04:21,088
to take a selfie with it and go on
LinkedIn and look absolutely fabulous.

1118
01:04:21,738 --> 01:04:24,798
Build a robo advisor
with Python from scratch.

1119
01:04:24,868 --> 01:04:25,508
Thank you guys.