Be Wary of Serial Correlation
MIT’s Andrew Lo:
The key concept here, developed by MIT professor and noted hedge-fund theorist Andrew Lo, is “serial correlation.” Simply put, serial correlation is the degree to which each month’s returns in a fund mirror the results of the month before. A fund that returns the exact same amount every month is perfectly serially correlated. Madoff’s returns were strikingly consistent month after month, year in and year out. That kind of performance—a nice, smooth line going up no matter what the market does—is a really good sign that you should look more closely.
The extraordinary thing that Lo does in the third chapter of his book Hedge Funds, published earlier this year, is to demonstrate mathematically that an excessive degree of serial correlation is a powerful indicator that the holdings of a fund aren’t being reported realistically. What Lo shows from the pattern of historical returns in hedge-fund databases is that when funds’ returns grow too consistent, it is a sign that the investments are either very hard to value accurately and the returns are just guesses, or, worse, that they’ve been manipulated in a way that smoothes them artificially. What Lo creates is a mathematical model for judging what “looks too good to be true.” Lo’s work turns a lot of the conventional thinking about what’s safe on its head. It shows that the evenness that investors have traditionally been taught indicates safety and reliability can actually be the best sign risk is being hidden or that the data are unreliable.
>
Source:
Madoff Dilemma: How Can You Spot A Wall Street Crook?
Mark Gimein
Big Money, Friday, December 12, 2008 – 3:57pm
http://www.thebigmoney.com/articles/judgments/2008/12/12/madoff-dilemma
Hedge Funds
Andrew W. Lo
Princeton University Press (May 18, 2008)
December 12th, 2008 at 6:15 pm
“looks too good to be true.”
Just more proof that everything you need to know, you learned in Kindergarten.
Mine, yours; right, wrong; it’s mostly just common sense isn’t it?
December 12th, 2008 at 6:17 pm
Serial Correlation… what a great tip off to a Ponzi scheme. If the results seem too good to be true, …
Who is beating the so-called professional hedge fund managers in the great Plunge of 2008?
It seems that some investors are beating the pros, here is a live test complete with real-time trades.
StrategicGrowthModel.blogspot.com
December 12th, 2008 at 6:20 pm
“It shows that the evenness that investors have traditionally been taught indicates safety and reliability can actually be the best sign risk is being hidden or that the data are unreliable.”
Volatility does NOT equal risk!!! Past volatility can be measured, but investment risk can never be known, even in hindsight.
December 12th, 2008 at 6:33 pm
While apparently not criminal I was always amazed how Cisco beat estimates every quarter by exactly 1 cent for years during the 90’s.
December 12th, 2008 at 6:37 pm
And GE under Welch. Same as Cisco but even more so…
Wasn’t it Bill Miller who said that the most overrated thing out there is consistent earnings growth?
December 12th, 2008 at 6:43 pm
Weren’t Lehman earnings along this line? I haven’t looked back at the EPS #’s. I wonder…
December 12th, 2008 at 6:44 pm
Patrick,
It’s somewhat easy if you have a positive cash flow. Pay less debt show more earnings.
That game works until recession or other stress comes.
December 12th, 2008 at 6:45 pm
Yay, something from grad school appears out in the real world! Uh oh, what are these things… people… back into the cave. *grumble grumble*
Moral of the story, if it’s too good to be true it is. Let that be repeated forever and ignored an equal amount.
December 12th, 2008 at 7:07 pm
It really is amazing, Lo has long done quality work, accessible to anyone with, the dimmest of, Spirit of Inquiry & a Library Card.
I’ll just guess it goes to show that there’s a grand difference between Professionals and ‘Pros’..
December 12th, 2008 at 7:11 pm
For what it’s worth in this connection a couple of my Financial industry contacts tell me the financial and legal elite of the street are in a state of total shock over the Madoff matter. Mainly because a hell of a lot of them have got all their investments in the hands of this guy. The other word out there is that quite a lot of informed people and organizations, including some at the SEC, have suspected for years that all was not well in the state of denmark with this guy but no one has blown the whistle. No doubt it’s all gooing to come tumbling out.
December 12th, 2008 at 7:20 pm
Watching the car crash known as CNBC today. Dylan prattling on about how, ah, could this, ah, happen, where were the regulators, in other words, ah….
This from the network that lionized Greenspan (remember the auction of the painting as Erin gushed about The My-es-tro? Hurl.). And Jack Welch (as earlier noted) is the Godhead Himself. Never mind that he may have just caught a helluva wave. Even now GE almost never misses by even a penny. Ludicrous.
How about Blankfein? Only he was smart enough to avoid subprime. Or was he? It’s the stinking cult of personality time and again. It’s why Thain actually thinks he deserves another dozen million. What a filthy bunch.
Despite what the gitbrained Kudlows and Caruso-Cabreras say, these clowns don’t deserve gazillions, because they aren’t substantially smarter 0r more insightful than their peers. Yes they’re bright and they work hard, but luck and good timing are just as important.
Oh, sweet fancy Moses, here’s Gasparino on Nightly News. There goes dinner.
December 12th, 2008 at 8:00 pm
It’s mostly the “fat cats” who took losses.
December 12th, 2008 at 8:05 pm
Now Citadel is barring redemptions…liquidity disappearing left and right…be ready for 500 point up days and yes, maybe a 1000 point down day coming.
BTW, not even asking a lot of people, I found a guy who’s relative lost 6 Million with Made-off.
December 12th, 2008 at 8:11 pm
Could the Mets now be bankrupt?
http://www.nydailynews.com/news/ny_crime/2008/12/12/2008-12-12_mets_owner_fred_wilpon_may_have_been_big.html
December 12th, 2008 at 8:22 pm
I worked as an actuarial student in the early 90s for a LARGE mutual life insurance company (which has since de-mutualized).
One of my bosses was a major player in putting out the company financials.
Since we didn’t need to conform to GAAP, this guy was OBSESSED with “levelizing” earnings, and we came up with all kinds of ways to obtain “serial correlation”.
December 12th, 2008 at 10:43 pm
msherwood1
I studied math. My boss didn’t. One day he barged into my office, totaly alarmed, holding a report for actuaries (or consultants as they are called) announcing: “Your fund is too risky”.
I tried to explain that, since I was managing a sectorial fund, all my stocks were correlated so his numbers meant nothing.” He did not like my reply and asked me to fix the problem.
Conclusion: we’re led by morons.
December 13th, 2008 at 12:40 am
Given these findings, why are so-called institutional investors so freaking obsessed with obtaining “smooth” returns with low volatility? They’re supposed to be smart and understand the difference between risk and volatility no?
December 13th, 2008 at 3:30 am
For what it’s worth in this connection a couple of my Financial industry contacts tell me the financial and legal elite of the street are in a state of total shock over the Madoff matter. Mainly because a hell of a lot of them have got all their investments in the hands of this guy. The other word out there is that quite a lot of informed people and organizations, including some at the SEC, have suspected for years that all was not well in the state of denmark with this guy but no one has blown the whistle. No doubt it’s all gooing to come tumbling out.
Wow! Talk about irony. For years the keystone cops were told to look the other way and they did. They got so good at it that they started to look the other way in the wrong places. The actually failed to protect their paymasters. The dog that couldn’t hunt finally learned how not to hunt too well! I never knew greed to have such a nasty, stinging and ironic tail but this is definitely one for the moral of the story book.
December 13th, 2008 at 3:46 am
@danm
Conclusion: we’re led by morons.
Make it idiot proof and someone will come along and build a better idiot….or they’ll just have a kid who will learn his gutter fighting from a gold plated tutor
@Francois
Given these findings, why are so-called institutional investors so freaking obsessed with obtaining “smooth” returns with low volatility? They’re supposed to be smart and understand the difference between risk and volatility no?
Volatility is for peasants
Pardon me folks for seeming to dance on a few freshly dug graves here but the market bullies have been dancing on mine for years and I’m not even dead yet. I can’t help but to think that events like this are going to deposit a little more respect into the account of Mr. Market(or is Mr. Market another lie dredged up by our elite but benevolent ‘leaders’?). When he shows up at the afternoon tea like the Godfather and machine guns everyone no matter who their tailor he begins to again assert control over what was supposed to be his in the first place
December 13th, 2008 at 9:25 am
Chiming in:
Yes, the idea that ‘volatility’ (variance) is an accurate measure of ‘risk’ is part of the religion of stock prices as statistics. It depends, as does the entire thesis, on the assumption that stock price reflects all actual forces at play (all the time) in a company’s performance. It should be obvious that an excessively smooth rate of return is an indicator of human intervention — only humans create straight lines. That this is some kind of ‘discovery’, reflects less than positively on the entire industry of mathematical finance and economics. Risk is in fact un-knowable, there is no metric for it, because we don’t have multiple copies of the real world around with which to compare outcome A vs. outcome B. The entire super-structure of physical theory, rooted, as the founders of quantum mechanics told us at length, in the procedures used in a lab, cannot, for that reason, simply be transfered to an entirely different arena.
I agree with DanM. The world has embraced some special kind of lunacy by asserting that ‘managers’, and ‘owners’ don’t actually need to know anything about that which they manage. It should be obvious (tee hee) that you can sort of get away with this when the company you own/manage makes staples (though in fact traditional companies, real companies, were owned and run by people who actually did know their business), but there just might be a teeny-weeny little bit of ‘not so much’ when the ‘product’ is the mathematical manipulation of markets. Of course this terribly obvious insight is subject to a caveat: depends what the game is. The assumption that the managers are playing a game whose goal is increasing asset value for ‘owners’ (investors) happens to be a terribly convenient thing for managers to have owners believe, particularly if the actual game is simply to increase the numbers in the managers personal accounts. Almost all criticism of both the financial and political world is rooted in the assumption that the major actors are engaging in an up and up game, the game we think they are supposed to be playing. It should be fairly obvious by now that this assumption is certainly invalid in specific cases, and may be invalid more widely.
December 13th, 2008 at 9:42 am
Question: is playing a con game reflected in price? Mabye someday, mabye tomorrow.
December 13th, 2008 at 9:42 am
VoiceFTW:
w/ this: “Almost all criticism of both the financial and political world is rooted in the assumption that the major actors are engaging in an up and up game, the game we think they are supposed to be playing. It should be fairly obvious by now that this assumption is certainly invalid in specific cases, and may be invalid more widely.”
nice point. needs to re-read, thought about, and further contemplated.
“Almost all criticism of both the financial and political world is rooted in the assumption that the major actors are engaging in an up and up game, the game we think they are supposed to be playing. It should be fairly obvious by now that this assumption is certainly invalid in specific cases, and may be invalid more widely.”
December 13th, 2008 at 9:57 am
DavidB Says: “… he begins to again assert control over what was supposed to be his in the first place”
good one … kinda like
resistance is futile – you will be assimilated
December 13th, 2008 at 10:43 am
It should be obvious that an excessively smooth rate of return is an indicator of human intervention — only humans create straight lines.
I thought it meant you had the blessing of the Fed
resistance is futile – you will be assimilated
….or broke (;
December 13th, 2008 at 12:58 pm
I await Lo’s sequel… haha…. to ‘Hedge Funds’ which hopefully will examine the serial increase in government debt since on near-surplus event during those horrible Clinton years.
I miss that laugh line “…You may not like Bush, but at least his done a good job with the economy…” anymore.
December 13th, 2008 at 1:48 pm
hmmm…. my FDIC insured CDs are looking might suspicious right now.
December 13th, 2008 at 6:06 pm
The SEC are a bunch of Cox Suckers.
December 13th, 2008 at 9:01 pm
Notice that since the early to mid-80’s the Dow Jones Industrial Average, Nasdaq, and the S&P 500 may suggest serial correlation. Someone should do some mathematical analysis on this. I’d be interested to hear comments on this. As Ritholtz suggests, we’re looking for a nice smooth line, with consistent almost unbelievable gains. From the 80s on, if you throw out the bubbles in each of these graphs, it looks very much like a nice pretty line extending to the sky. Year on year growth, as any financial adviser at any mutual fund or retirement investing firm will say, can be expected to be in the 10% range over the long haul – because the stock market always goes up.
An interesting note: the invention of the 401(k) in 1978 and almost half of U.S. corporations offering them by around 1983 correlates to the beginning of the last roughly 30 years of increases in the DJI. The universalization of 401(k)’s is today easily apparent.
Are all the U.S. workers contributing to 401(k)s and IRAs just a part of one big Ponzi Scheme?
December 14th, 2008 at 11:04 am
Beside looking at so called serial correlation, there is a more down to earth way for the math & stat less skilled.
If you want to know if there is a simple function that estimates the next return in a (long) time series,
that is you wonder if:
r_{n+1} = f(r_n), or
tomorrow = f(today)
you simply plot (r_n, r_{n+1}) for the entire time series.
In case of financial thingelings there should be ‘enough chaos’ because there are always many unforeseen things; if the points are too close together you should investigate further.
December 14th, 2008 at 4:28 pm
“In case of financial thingelings there should be ‘enough chaos’ because there are always many unforeseen things; if the points are too close together you should investigate further.”–Reinko
on the flip side, remember: “Stability breeds Chaos.” for further, see: Minsky, et al.