Saturday, February 12, 2011

Information Entropy And The Stock Market Revisited

Relatively recently, FLG focused on information entropy as a way of understanding the efficient market hypothesis and stock market fluctuations.

Today, FLG saw that Withywindle recommended a biography of Keynes by Robert Skidelsky.  A bit of googling revealed this article in which Skidelsky writes:

In my view, Keynes’s major contribution to economic theory was to emphasize the “extreme precariousness of the basis of knowledge on which our estimates of prospective yield have to be made.” The fact of their ignorance forces investors to fall back on certain conventions, of which the most important are that the present will continue into the future, that existing share prices sum up future prospects, and that if most people believe something, they must be right.
This makes for considerable stability in markets as long as the conventions hold . But they are liable to being overturned suddenly in the face of passing bad news, because “there is no firm basis of conviction to hold them steady.” It’s like what happens in a crowded theater if someone shouts “Fire!” Everyone rushes to get out. This is not “irrational” behavior. It is reasonable behavior in the face of uncertainty. In essence, this is what happened last autumn.

2 comments:

nadezhda said...

Keynes was an active investor/trader/speculator who both won and lost a great deal of money, so his insights on market dynamics are often wryly spot-on.

As for EMH, don't forget the timing problem, which limits the ability of the market to fully incorporate information. via Brad DeLong, here's a nice post from Peter Dorman at EconoSpeak. As Keynes famously noted, "Markets can remain irrational far longer than you or I can remain solvent." It applies not only to bubbles but to "depressed" or over-sold markets.

nadezhda said...

I think you'll also like this post by Rajiv Seth on a conference last week at Columbia titled "Heterogeneous Expectations and Economic Stability".

 
Creative Commons License
This work is licensed under a Creative Commons Attribution-No Derivative Works 3.0 United States License.