Conditional betas and the price of risk in a thin asset market : A sensitivity analysis

Abstract

This paper examines the sensitivity of tests of the Sharpe-Lintner Capital Asset Pricing Model (CAPM) to different estimation methods and asset return samples in a thin European asset market, i.e. the Finnish asset market. A time-varying-parameter model is introduced as an altemative to the static market model. We run a regression over a pooled data set in addition to the second-pass Fama-McBeth regressions. Our tests are carried out with four asset specific samples. In every case, cross-sectional OLS estimation of the betas leads to the rejection of the mean-variance efficiency of the market index. The price of market risk is statistically significant, but negative. Our tests on the time-varying betas indicate just the opposite. We are. not able to reject the mean-variance efficiency of the market index in any of the samples. The price of market risk is positive and statistically significant for the stock return data set that most closely resemble the normal distribution.