Fundamental market ratios and stock return predictability in Karachi stock exchange

Abstract

In this paper, an attempt has been made to examine the Karachi Stock Exchange by reviewing theoretical and empirical literature on efficient markets and predictive power of fundamental market based ratios; in particular: Dividend Yield, Price Earnings, Book to Price ratio and Market Capitalization (Size). The work aims to conduct an in depth investigation into the relationship that exists between stock returns and the four fundamental market ratios using the Ordinary Least Square Regression Model. A sample size of 32 KSE listed companies has been taken, gathering monthly data over a period of 10 years (2000 – 2010). Under the OLS Regression mode, a total of three tests are run: fixed-effects, random-effects and pooled-effects. The fixed and random tests have produced similar results; however, the pooled test has shown different outcomes. Both the fixed-effects test and the random-effects test find only the P/E ratio having a statistically significant positive relationship with stock returns. Whilst the pooledeffects tests which identifies all three market fundamentals: Dividend yield, Price earnings and Size as having statistically significant relationships with stock returns. All of the tests have obtained a statistically insignificant relationship between stock returns and market to book ratio. The model also identifies a mystery variable that shares a robust positive relationship with stock returns in both the random and pooled tests. However, it remains unexplained in the paper. Market beta also has a considerably strong positive relation to stock returns but it has not been considered as an important independent variable in the paper. The impact of both the mystery variable and market beta is strong in the test results, which alters the ultimate outcome of the tests, the R2 value. Where, individually the chosen market fundaments (DY, P/B and Size – with the exception of P/E) do not hold a statistically strong association with stock returns, together with the added strength of mystery variable and market beta, the R2 value of the tests rises, the tests end up explaining 27 percent of the variations in stock returns through the market fundamentals. The value of R2 remains consistent in all of the three tests.