Losing At the Casino

For some time, but until now unfollowed, I've harbored the notion that MCMC (Markov Chain Monte Carlo) modeling might be suspect in The Great Recession. Turns out, I'm not the only one.

Skinny Tails

Monte Carlo simulation is a convenient statistical tool by which an analyst can make inferences over the probability of rare events. Basically, the method takes a model with its estimated parameters and simulates possible sample paths under some distributional assumptions on the errors. The technique is particularly useful for understanding the tail risk of distributions. The method has been widely adopted by banks, investors, and financial advisors. And, because the models failed so horribly in the current downturn, the method is now under attack.

Ya think????

Felix Solomon's take

So the lesson here, I think, is mostly that stock-market tails are fat. But there is a sub-lesson, too, which is that Monte Carlo simulations can be very dangerous, if they're implemented by people who don't know what they're doing. Including the quants at Moody's.

A 2007 critique (who knew anyone refused the Flavour Aid?)

The amount of derivatives in play is in the hundreds of trillions of dollars all bet on Monte Carlo Simulation. Wow!
...
It sure seems like a hell of a lot is riding on the Monte Carlo Simulation of CDOs as well as the parameters and assumption that feed into the model. If and when a Credit Event occurs that rifles through multiple CDO tranches, the guarantors will be about as well capitalized to handle the guarantees as is Madame Merriweather's Mudhut Malaysia. Do I smell a government bailout coming up?

Ya think???

Practical magic?

The downside of Monte Carlo is that it is more trusted than historical data. This misplaced trust is rooted in the idea that if a person has no historical data, then the Monte Carlo forecast can be anything that is believed plausible. For example, the phrase: "Contingency was determined by running Monte Carlo" is equivalent to "PI was determined by using Monte Carlo." Monte Carlo could render reasonable results for each value. They both will not be precise answers because Monte Carlo is an approximation, and the results for both could be completely wrong.

The issue with such methods is that they view human actions as Brownian motion (really!). While reasonable for physical processes, not so much for finance, where the rules are made up by humans and can be changed at a whim. That's why we got our Great Recession: humans changed the rules faster than the data streams could reveal the changes. The analysts made their decisions under the assumptions that Tomorrow will look pretty much like Today, which in its turn looks pretty much like Yesterday; and, as with physical processes, that the platform was stable during measurement. The fact that median house price had come unstuck from median income wasn't relevant to the modellers or their models; to look in that direction wasn't in the self-interest of those engaged in the game. Alas, if you look at the data through a toilet paper tube, you'll not see that hyena licking his chops for dinner. You're dinner.

And, as past is prologue, comes reports that brokerage houses are luring retail investors into options trading. There's nothing more fun than playing with nitroglycerine whilst bungee jumping.

At Ameritrade, which has been the most aggressive, derivatives trades accounted for about 40 percent of all customer trades last year -- more than double what it was just five years ago. A vast majority of those trades were in options.

The growth has been a big help for the online brokers at a time when stock trading has fallen. The commission on the average options trade is more than twice that on the average stock trade, according to TD Ameritrade's former treasurer, Michael Chochon.
...
[H]e saw investors taking up options trading and "blowing up" on an almost daily basis. He said Ameritrade carefully tracked the risks its customers were taking but did not warn them until they were close to losing it all, if then.

If that sounds a bit like the Banksters grinding ever more, and expensive, fees from common depositors to make up for the loss of all those wonderful profits on subprime and such, well "yes I said yes I will Yes" (that's from a famous dirty book, btw).