Efficiency of the Philippine stock market

Abstract

The study examines the efficiency of the Philippine stock market using stock price movements during the period July 1987 to May 2004. Characterizing stock price movements as an AR(1) process with Laplace residuals, the statistical evidence supports the hypothesis that the Philippine stock market is weak-form efficient. An examination of major events that could plausibly affect share prices and large price movements from an event study perspective indicates fairly rapid absorption by the market of information, except in cases of extreme stress caused by political and economic shocks. Furthermore, factors other than information about fundamentals appear able to cause major share price movements. Given these, the support for the semistrong-form efficiency of the stock market is mixed.

Acknowledgements

The author is grateful to Roy Ybañez and Joel C. Yu for their valuable comments.

Notes

¹ The Phisix is a market-value-weighted index of 30 representative companies from different sectors of the local bourse. Data is from the Philippine Stock Exchange.

² Ordinary least squares (OLS), which is the best linear unbiased estimator even in the presence of non-normal errors, is used.

³ The first-order autocorrelation coefficient is also estimated using the least absolute deviation method which is the maximum likelihood estimator when the errors follow the Laplace df (McDonald, 1996). The result is not statistically different from the OLS estimate.

⁴ The Laplace distribution belongs to the family of elliptical distributions important in finance theory. See Johnson and Kotz (1970) for an extensive reference on this distribution.

⁵ Alternatively, Linden (2001) showed that the Laplace distribution can be derived from a conditional normal distribution with stochastic variance.

⁶ For events in the USA, the event day is the following day, Philippine time.

⁷ EDSA is the name of the major road where protestors against the government congregated.