Behavior of Asset Prices

EVIDENCE THAT THEY DO NOT BEHAVE RATIONALLY:A DISCUSSION OF THE ISSUE

A procedure which permits an accurate and timely prediction of common stock and bond prices has been sought since the inception of capital markets. The quest, thus far, has proved to be a largely futile effort. As the rewards to be derived from such a procedure are great, however, the quest has endured over the decades.

The Character of Stock Price Changes

Bachelier (1900) provided important early work on the problem of predicting stock price changes, reasoning that the expected value of a stock price at time t + 1, knowing the time series of all past prices, is simply the present price. The eventualresult of this work was the Random Walk Model.

Bachelier (1900) theorized that successive price changes are drawn from the same probability distribution. Based on a "fair game" characterization of the marketplace, it was assumed that price changes followed a random walk with a normal distribution. Therefore, assuming independent price changes on successive transactions, and with a process having finite variance, application of the Central Limit Theorem led to a conclusion that, after many transactions, the limiting distribution must be normal. At the time Bachelier wrote, very little empirical evidence could be marshaled to support these claims. Thus, the work lay dormant for 50 years, until analysis o

. . .
al Average stocks. The size of the data set is important, because more points in the relatively sparse but excessively high tails (relative to a normal distribution) appear with larger data sets. It was concluded that a stable Paretian distribution was a better fit to the fat tailed data than the normal density. Price Distribution andTrading Volume________ Trading volume has been linked to price volatility. Ying (1966), Epps (1975), and Rogalski (1978), among others, found a correlation between the magnitude of daily price changes and the trading volume relative to an average volume for the stock. Bey and Pinches (1980) found that about 25 percent of stocks were heteroscedastic (changing variances). Belkaoui (1977) found that 91 percent showed evidence of heteroscedasticity. Not only does the variance change with time, but it is also evidently related to the price level of the stock itself as the square root of the stock price. Christie (1982) showed that the volatility increases with a firm's financial leverage, and that it is sensitive to the prevailing interest rate. Compound Events Models As investigations indicated that distributions were characterized by both "fat tails" and heteroscedasticit
. . .

Some common words found in the essay are:
Brealey Myers, Limit Theorem, Model Bachelier, Granger Morgenstern, DISCUSSION ISSUE, Changes Bachelier, random walk, price changes, stock price, seligman 1984, runs test, movements data set, recurring pattern, data set, stock prices, bachelier 1900, movements data, stock price changes,
Approximate Word count = 2655
Approximate Pages = 11 (250 words per page)

More Essays on Behavior of Asset Prices