II  The Wall Street Math Hustle
II  The Wall Street Math Hustle
Last edited by swaption on Tue Nov 06, 2018 2:30 pm, edited 1 time in total.
Re: II  The Wall Street Math Hustle
I liked this part:swaption wrote: ↑Tue Nov 06, 2018 11:23 amHave my views on this, interested in others.
https://www.institutionalinvestor.com/a ... h%20Hustle.
Real math is painfully precise. (Watch the movie Hidden Figures if you didn’t already know that.) Investment finance math is sales math. It doesn’t matter that you don’t need it or that everybody interprets it in their own way. You’re using it to impress clients.
Bogleheads® wiki 
Investing Advice Inspired by Jack Bogle
Re: II  The Wall Street Math Hustle
I don't think the dataset was adjusted for survivorship bias. H
"whenever there is a randomized way of doing something, then there is a nonrandomized way that delivers better performance but requires more thought" ET Jaynes
 nisiprius
 Advisory Board
 Posts: 37604
 Joined: Thu Jul 26, 2007 9:33 am
 Location: The terrestrial, globular, planetary hunk of matter, flattened at the poles, is my abode.O. Henry
Re: II  The Wall Street Math Hustle
As best I can tell in a quick skim, I think it's right, and a good article.
I quit believing in Wall Street math some years ago, when I noticed that nobody was even trying to explain or justify the use of parametric statistics like "Tstats", when nonparametric tests have been available for decades... other, of course, than the venal justification that parametric tests are more likely to show "statistical significance."
I quit believing in Wall Street math some years ago, when I noticed that nobody was even trying to explain or justify the use of parametric statistics like "Tstats", when nonparametric tests have been available for decades... other, of course, than the venal justification that parametric tests are more likely to show "statistical significance."
Annual income twenty pounds, annual expenditure nineteen nineteen and six, result happiness; Annual income twenty pounds, annual expenditure twenty pounds ought and six, result misery.

 Posts: 2157
 Joined: Fri Aug 06, 2010 3:42 pm
Re: II  The Wall Street Math Hustle
I enjoyed reading this article and have finally encountered someone who is more skeptical about Wall Street than I am.
Garland Whizzer
I think it's important to hear both sides of complex investing questions like factors versus cap weight, how to achieve return efficient diversification, etc.. Wiggins and Edesess present a counter argument to complex strategies to achieve investment goals. They suggest that the theories on which these strategies are based tend to mistake random market noise for true signal in mathematical backtesting. I'm not sure who's right on this point but it's worth a read.For many years, alpha was the Holy Grail of investment and almost all money managers claimed they could produce it. Then, under a Jupitersized weight of evidence that no investment manager — not mutual funds, not hedge funds — could reliably produce alpha, investment managers stopped laying outright claims to it.
Now, for many, there is a new Holy Grail: diversification.
Unable as a group to show that they can get returns, the industry changed the subject by shifting the focus of investing from return to risk.
Garland Whizzer
Re: II  The Wall Street Math Hustle
Your link contains UTM tracking codes that tell Google and other web trackers where you got the link (https://en.wikipedia.org/wiki/UTM_parameters). You should edit your link to remove the tracking info. Here is the direct link: https://www.institutionalinvestor.com/a ... athhustle .
As to the article, I believe that telling people what to invest in will always be significantly influenced by marketing. Using fundamental math is useless against the forces of marketing. As a Boglehead, I just give up trying to understand the mathematics behind what I should be investing in and I'll just be satisfied with average returns.
 triceratop
 Moderator
 Posts: 5831
 Joined: Tue Aug 04, 2015 8:20 pm
 Location: la la land
Re: II  The Wall Street Math Hustle
Hahaha . As someone who actually uses math and reads/understands/disbelieves/occasionally writes proofs the idea that what the financial industry does is anywhere close to the level of mathematical proof or rigor expected in actual math is just hilarious....indeed, the requirement — today is for firms to use scientific language and notation to nourish the idea that they’ve proved mathematically that there’s a way to systematically beat the market.
"To play the stock market is to play musical chairs under the chord progression of a bidask spread."
Re: II  The Wall Street Math Hustle
Pure math (math without numbers) is difficult.triceratop wrote: ↑Tue Nov 06, 2018 2:01 pmHahaha . As someone who actually uses math and reads/understands/disbelieves/occasionally writes proofs the idea that what the financial industry does is anywhere close to the level of mathematical proof or rigor expected in actual math is just hilarious....indeed, the requirement — today is for firms to use scientific language and notation to nourish the idea that they’ve proved mathematically that there’s a way to systematically beat the market.
Last edited by Tycoon on Wed Nov 07, 2018 1:59 pm, edited 2 times in total.
Be still, sad heart! and cease repining; 
Behind the clouds is the sun still shining; 
Thy fate is the common fate of all, 
Into each life some rain must fall, 
Some days must be dark and dreary. 
Longfellow
Re: II  The Wall Street Math Hustle
Original poster here. In my opinion, a bunch of garbled, barely coherent BS. Sure, there are a couple of ok ideas. But he uses his math quite selectively, throwing neat statistics around, and in some case throwing nothing around. Take the following passage:
This is rationalization masquerading as real analysis. Managers can only be as smart as the money that comes to their funds, so perhaps systemically many essentially chase the money that chases the returns. Not saying this is right or wrong, but the simplistic answer of "the "managers would have gotten it by now" is just ridiculous. Sure, if this were a world where managers could be myopically focused on long term returns, immune to the economic realities of attracting capital, then maybe he'd have a point. But there are agency and behavioral factors at work here, while maybe open to debate, they can't be completely dismissed.If it were that easy, active managers would be beating their indexes year after year because by now we’re sure they’ve gotten the memo about Rolf Banz’s 30yearold paper “proving” that smallcap stocks beat large caps, which, by the way, isn’t true: He forgot to transform monthly holdingperiod returns to log returns before running his regression.
 nisiprius
 Advisory Board
 Posts: 37604
 Joined: Thu Jul 26, 2007 9:33 am
 Location: The terrestrial, globular, planetary hunk of matter, flattened at the poles, is my abode.O. Henry
Re: II  The Wall Street Math Hustle
Is Wiggins the same person who pointed out that Fama and French use a seemingly arbitrary classification system, in which size gets a twocategory classification but value gets threeand the fact that the two factors are treated differently is never mentioned, let alone justified?
Annual income twenty pounds, annual expenditure nineteen nineteen and six, result happiness; Annual income twenty pounds, annual expenditure twenty pounds ought and six, result misery.
Re: II  The Wall Street Math Hustle
Thank for sharing, I agree with most everything in the article...
A statistic I found interesting (but not surprising a la Pareto distributions):
A statistic I found interesting (but not surprising a la Pareto distributions):
Loved this bit:... According to Hendrik Bessembinder, a finance professor at Arizona State University, the entire gain in the U.S. stock market since 1926 is attributable to the bestperforming 4 percent of listed companies, and the capweighted indexes captured all of it because they don’t rebalance.
... To the extent that this difference is large, an investor has put together securities that are intrinsically different from each other — but it doesn’t necessarily mean it’s more diversified than the marketcapweighted portfolio. Betting on both red and black at the same time in roulette delivers a mammoth diversification ratio. It’s also stupid.
Maximum diversification has a name and story so fat that, like yelling “Hey KoolAid!” investors come crashing through the wall. What they’re leaving out of the sales pitch is that the maximum diversification portfolio is always more concentrated relative to the marketcapweighted portfolio in terms of risk contribution.
"To achieve satisfactory investment results is easier than most people realize; to achieve superior results is harder than it looks."  Benjamin Graham
Re: II  The Wall Street Math Hustle
So in other words, small and value premiums may entail additional risk. Just seems like a bunch of bloviated stuff saying just that.JoMoney wrote: ↑Tue Nov 06, 2018 6:36 pmThank for sharing, I agree with most everything in the article...
A statistic I found interesting (but not surprising a la Pareto distributions):Loved this bit:... According to Hendrik Bessembinder, a finance professor at Arizona State University, the entire gain in the U.S. stock market since 1926 is attributable to the bestperforming 4 percent of listed companies, and the capweighted indexes captured all of it because they don’t rebalance.... To the extent that this difference is large, an investor has put together securities that are intrinsically different from each other — but it doesn’t necessarily mean it’s more diversified than the marketcapweighted portfolio. Betting on both red and black at the same time in roulette delivers a mammoth diversification ratio. It’s also stupid.
Maximum diversification has a name and story so fat that, like yelling “Hey KoolAid!” investors come crashing through the wall. What they’re leaving out of the sales pitch is that the maximum diversification portfolio is always more concentrated relative to the marketcapweighted portfolio in terms of risk contribution.
 Taylor Larimore
 Advisory Board
 Posts: 27827
 Joined: Tue Feb 27, 2007 8:09 pm
 Location: Miami FL
90year study: Largecaps vs. Smallcaps.
Bogleheads:
I wonder how many smallcap advocates will read this:
Taylor
I wonder how many smallcap advocates will read this:
Best wishes.According to a 90year study by Bessembinder, when buyandhold portfolios of stocks are sorted by beginning market cap and held for a decade, largecap portfolio returns are higher (153 percent) than the mean returns for smallcap portfolios (97 percent)
Taylor
"Simplicity is the master key to financial success."  Jack Bogle

 Posts: 184
 Joined: Mon Feb 20, 2012 8:16 pm
 Location: San Diego
Re: II  The Wall Street Math Hustle
I thought it was good, and is a reminder of what has been proven and stated many times: markets are efficient. The best, simplest and most mathematically pure way to invest is marketcap based indexing. Despite Wall Street's continued marketing ponzi schemes, the amount of success they've had in attracting assets to things like smart beta or equal weight index funds has been marginal compared to market cap weighted index funds.
But the mathematical and practical beauty of marketcap based indexing is a message that needs to be consistently repeated, so I applaud the article, and I really like its concluding paragraph.
But the mathematical and practical beauty of marketcap based indexing is a message that needs to be consistently repeated, so I applaud the article, and I really like its concluding paragraph.
 triceratop
 Moderator
 Posts: 5831
 Joined: Tue Aug 04, 2015 8:20 pm
 Location: la la land
Re: 90year study: Largecaps vs. Smallcaps.
The word "portfolio" is doing a lot of work in that sentence. The original article, and you, are misreading (or not reading) the original research study by Bessembinder which finds in Do Stocks Outperform Treasury Bills?:Taylor Larimore wrote: ↑Tue Nov 06, 2018 8:22 pmBogleheads:
I wonder how many smallcap advocates will read this:
Best wishes.According to a 90year study by Bessembinder, when buyandhold portfolios of stocks are sorted by beginning market cap and held for a decade, largecap portfolio returns are higher (153 percent) than the mean returns for smallcap portfolios (97 percent)
Taylor
I very much think that the study doesn't show what you think it is showing. Skewness in individual stock returns is not the same as returns in market cap portfolios targeting a given decile.Bessembinder wrote:Despite the fact that small firms deliver higher mean monthly returns as compared to large, the data reported on Table 3A show a distinct pattern by which small stocks display more return skewness and a higher frequency of underperformance relative to benchmarks.
"To play the stock market is to play musical chairs under the chord progression of a bidask spread."
 alpine_boglehead
 Posts: 277
 Joined: Fri Feb 17, 2017 9:51 am
 Location: Austria
Re: II  The Wall Street Math Hustle
It's a bit lengthy to read and yes, it's a wild mix of ideas, but there's quite some gems already pointed out by the other posters.swaption wrote: ↑Tue Nov 06, 2018 2:48 pmOriginal poster here. In my opinion, a bunch of garbled, barely coherent BS. Sure, there are a couple of ok ideas. But he uses his math quite selectively, throwing neat statistics around, and in some case throwing nothing around. Take the following passage:
This is rationalization masquerading as real analysis. Managers can only be as smart as the money that comes to their funds, so perhaps systemically many essentially chase the money that chases the returns. Not saying this is right or wrong, but the simplistic answer of "the "managers would have gotten it by now" is just ridiculous. Sure, if this were a world where managers could be myopically focused on long term returns, immune to the economic realities of attracting capital, then maybe he'd have a point. But there are agency and behavioral factors at work here, while maybe open to debate, they can't be completely dismissed.If it were that easy, active managers would be beating their indexes year after year because by now we’re sure they’ve gotten the memo about Rolf Banz’s 30yearold paper “proving” that smallcap stocks beat large caps, which, by the way, isn’t true: He forgot to transform monthly holdingperiod returns to log returns before running his regression.
Your point that managers just cater to the taste of investors may be right in some situations. On the other side there's banks, brokers and advisers pushing overpriced investments on to clueless (retail) investors. And the article is right that the tune of the sales pitches has become more sophisticated.
Just saw an ad of a local bank that now provides "diversified investments via ETFs". They created a fund that holds ETFs, put the "diversification" marketing tag on it and charge 1.35% expense ratio. Buyer beware.
Re: II  The Wall Street Math Hustle
There is one small problem with the paper, as far as I can tell. The two curves shown in the paper in the figure entitled "Slippery Correlation Ramp" which I recreated using R in the code below don't have a negative correlation at all. In fact they are strongly positively correlated. I am mostly in agreement with the overall thesis, however.
This chart shows two negatively correlated portfolios, correlation coefficient is 1
This chart shows two positively correlated portfolios  Correlation coefficient is 0.98
R code is here
This chart shows two negatively correlated portfolios, correlation coefficient is 1
This chart shows two positively correlated portfolios  Correlation coefficient is 0.98
R code is here
Code: Select all
test.sine < function(slope, shift, x){
sin(x+shift) + slope*x
}
x < seq(0,8*pi,0.001)
jpeg(filename="corplot1.jpg")
# Draw the curves with zero slope.
matplot(x, cbind(test.sine(slope=0,shift=0,x), test.sine(slope=0,shift=pi,x)), main='Two negatively correlated portfolios')
dev.off()
# Now add a common slope as in the paper.
jpeg(filename="corplot2.jpg")
matplot(x, cbind(test.sine(slope=1,shift=0,x), test.sine(slope=1,shift=pi,x)), main='Slope induces positive correlation')
dev.off()
# Compute the correlation if slope is zero.
cor(test.sine(slope=0, shift=0,x),test.sine(slope=0,shift=pi,x))
# Correlation will be 1
cor(test.sine(slope=1, shift=0,x), test.sine(slope=1,shift=pi,x))
# Correlation is 0.9818
Re: II  The Wall Street Math Hustle
Wow! Holy c**p! Their prospectus should make for interesting reading. You could set up as an investment advisor in Austria and undercut them at 1% by buying facebook ads!!!alpine_boglehead wrote: ↑Wed Nov 07, 2018 2:32 pmJust saw an ad of a local bank that now provides "diversified investments via ETFs". They created a fund that holds ETFs, put the "diversification" marketing tag on it and charge 1.35% expense ratio. Buyer beware.
 alpine_boglehead
 Posts: 277
 Joined: Fri Feb 17, 2017 9:51 am
 Location: Austria
Re: II  The Wall Street Math Hustle
Would be an idea  this "ETF fund" already has 15 million AUM, so a similar scheme with 1% ER less 0.2% ETF fees would make you 120k per year. A look at the prospectus shows 32% in S&P 500 ETFs, 14% MSCI Europe ETF, and so on ... they're fleecing investors who could just hold a total stock and total bond ETF.edgeagg wrote: ↑Wed Nov 07, 2018 4:43 pmWow! Holy c**p! Their prospectus should make for interesting reading. You could set up as an investment advisor in Austria and undercut them at 1% by buying facebook ads!!!alpine_boglehead wrote: ↑Wed Nov 07, 2018 2:32 pmJust saw an ad of a local bank that now provides "diversified investments via ETFs". They created a fund that holds ETFs, put the "diversification" marketing tag on it and charge 1.35% expense ratio. Buyer beware.
Re: II  The Wall Street Math Hustle
Amazing!! Time to set up the Alpine Boglehead Fund!!! Facebook ads aren't that expensive. You recall that WealthFront was doing something similar until their latest pivot to smart beta.....alpine_boglehead wrote: ↑Thu Nov 08, 2018 1:21 pmWould be an idea  this "ETF fund" already has 15 million AUM, so a similar scheme with 1% ER less 0.2% ETF fees would make you 120k per year. A look at the prospectus shows 32% in S&P 500 ETFs, 14% MSCI Europe ETF, and so on ... they're fleecing investors who could just hold a total stock and total bond ETF.
Re: II  The Wall Street Math Hustle
edgeagg wrote: ↑Wed Nov 07, 2018 3:07 pmThere is one small problem with the paper, as far as I can tell. The two curves shown in the paper in the figure entitled "Slippery Correlation Ramp" which I recreated using R in the code below don't have a negative correlation at all. In fact they are strongly positively correlated. I am mostly in agreement with the overall thesis, however.
This chart shows two negatively correlated portfolios, correlation coefficient is 1
This chart shows two positively correlated portfolios  Correlation coefficient is 0.98
R code is hereCode: Select all
test.sine < function(slope, shift, x){ sin(x+shift) + slope*x } x < seq(0,8*pi,0.001) jpeg(filename="corplot1.jpg") # Draw the curves with zero slope. matplot(x, cbind(test.sine(slope=0,shift=0,x), test.sine(slope=0,shift=pi,x)), main='Two negatively correlated portfolios') dev.off() # Now add a common slope as in the paper. jpeg(filename="corplot2.jpg") matplot(x, cbind(test.sine(slope=1,shift=0,x), test.sine(slope=1,shift=pi,x)), main='Slope induces positive correlation') dev.off() # Compute the correlation if slope is zero. cor(test.sine(slope=0, shift=0,x),test.sine(slope=0,shift=pi,x)) # Correlation will be 1 cor(test.sine(slope=1, shift=0,x), test.sine(slope=1,shift=pi,x)) # Correlation is 0.9818
The comment above applies only because the author of the comment appears to be correlating the series of PRICES of the assets. If you correlate the series of RETURNS the correlation coefficient remains minus one.
Re: II  The Wall Street Math Hustle
The comment applies only because the author of the comment appears to be correlating the series of PRICES of the assets. If you correlate the series of RETURNS the correlation coefficient remains minus one.