Random Walks and Fractals

There is an article by Benoit Mandelbrot and Nassim Nicholas Taleb, which is a superb explanation of the deficiencies in modern financial theory "How the Finance Gurus Get Risk All Wrong". It is also a reminder to all of us that we are captives to the ebbs and flows of the major academic trends of the time. As Keynes had said, " Practical men, who believe themselves quite exempt from any intellectual influence, are usually slaves of some defunct economist".

Despite its proven obsolescence, standard measures of risk are still holy cows in the corridors of modern finance. Power Laws have yet to make any dent in the Gaussian Distribution dominated world of modern finance. Classical economics, revolving around concepts of equlibrium and the utility maximizing rational economic man with access to perfect information, continues to hold sway in Eco 101 courses across the globe. Business Cycles and other major events continues to be explained by the exogenous shock theory. No where is this disconnect more stark than in the practical world of equity markets. Mandelbrot and Taleb says:

"Your mutual fund's annual report, for example, may contain a measure of risk (usually something called beta). It would indeed be useful to know just how risky your fund is, but this number won't tell you. Nor will any of the other quantities spewed out by the pseudoscience of finance: standard deviation, the Sharpe ratio, variance, correlation, alpha, value at risk, even the Black-Scholes option-pricing model. The problem with all these measures is that they are built upon the statistical device known as the bell curve. This means they disregard big market moves: They focus on the grass and miss out on the (gigantic) trees."

The same applies for our regard and the importance attached to stock analyst valuation of equities. The fundamental parameters necessary for valuation of any asset are its projected cash flows and the rate at which these flows are discounted, besides the other environmental factors like inflation, taxes etc. But there are no fail-proof methods for calculating the afore-mentioned two parameters. Despite this, stock valuation and stock picking are the accepted norms in modern finance theory. Instead of asset allocation and portfolio diversification strategies, investors are lured straight into the glitz of stock-picking. As they say "never have so many intelligent people been all chasing such completely false theories". For more on this, Henry Blodget has an excellent piece in the Slate, "Stop Picking Stocks - Immediately!".

Update 1
Nassim Nicholas Taleb excoriates the portfolio theory, law of large numbers, and the pervasive use of statistics in finance, while Harry Markowitz mounts a defence of MPT, distinguishing it from financial engineering.