Efficiency of a market could be measured in terms of the contribution of value variance and noise to the total return variance. This study adopts the model used by Damodaran (1993) to compute the value variance and noise for the stocks in the "A", "B1" and "B2" groups of the BSE and compares with similar cross-listed stocks at the NSE. The study is done using daily closing prices for the period from January 2002 to December 2004. The results show an increase in the contribution of noise to the total return variance as one moves from "A" group to "B1" to "B2" group for both the exchanges. Also the component of noise is more than value variance in the total return variances for stocks listed at the NSE when compared to similar stocks at the BSE.

In an efficient market, information is instantaneously reflected in prices. However, in actual market some time elapses before the participants undertake trades to make the market efficient. The changes in price on account of these trades could be brought about either by the informed traders or by noise traders. Thus, variance in prices would be the result of either variance in value trading (by the informed traders) or variance due to noise trading (by the noise traders). Intuitively, the changes in price in an efficient market would be brought about more by informed traders than noise traders. Thus, if one can segregate the variance in prices into the twin components of variance due to value and variance due to noise, then the ratio of these two variances for a given market would indicate whether one market is efficient as compared to the other market. A higher ratio of value variance to noise variance for a given market as compared to another market would indicate the market which is more efficient.

Valuation Errors of Cross-listed Stocks in BSE and NSE, efficient, traders, market, stocks, trades, contribution, exchanges, account, elapses, information, instantaneously, listed, adopts, participants, reflected, component, segregate, Damodaran, Efficiency