Testing the capm in the Indian market, a A study that investigates the validity of the CAPM in Bombay Stock Exchange SENSEX30

This paper is designed to examine the validity of the CAPM model in the emerging markets. I took the Indian market to be the case in which we examine the applicability of this model and therefore I decided to perform the study on one of the biggest Indian markets; Bombay Stock Exchange. The SENSEX30 was chosen as the examined index and I performed the study on the 28 listed companies in the market (BSE30 or SENSEX30). I used weekly stocks’ returns for the period Nov04 to OCT09. To eliminate the measurement bias which will be incurred during the study, a window of 53 weeks was taken to regress the weekly returns of the listed stocks on the weekly returns of the SENSEX30 index at the same period, this will result in 53 betas for each stock in the first period, and then we started to move the window week by week. When testing the CAPM model for the whole five-year period hasn’t showed any strong evidence that support the validity of this model and in order to get better estimates, we divided the whole sample into 5 subsamples of one year each. We have examined three tests for each year of the whole 5 year sample and the results have shown some better estimates for some of the years but still did not support the CAPM hypothesis When running the non-linearity test, it was proven that the model explains the excess returns which-in return-supports the linear structure of the CAPM equation. This paper is solely based on the fact that in order for the model to be valid and strong academically, the alpha (the intercept) should equal to zero and the beta (the slope) should equal the excess returns on the market portfolio. This was a pure prediction of the CAPM, therefore we tested the above hypotheses but the results failed to prove or provide any evidence that coincide with the null hypotheses! The second part of this investigation was to examine the ability of the CAPM model to provide a non-linearity relationship between the return and betas. Conducting the test has shown that the expected return-beta relationship is linear. In fact, this paper has gone too far by also including a brief investigation over what’s called “the non-systematic test”. The idea behind this was to investigate whether the CAPM can include all the components of the stocks’ returns including the residual variance of the stocks. Our resultsbased on the test for the non-systematic risk-show that the residual risk has no effect on the expected returns of the listed stocks