Conventional tests of capital asset pricing model usually assume that β, a measure of the systematic risk, is stable over time. Nonetheless, empirical investigations generally find that β tends to be volatile over time. This paper formulates novel empirical models for testing the capital asset pricing model (CAPM) by allowing β to be drawn from two distinct regimes, or to be time varying. The econometric method via the Gibbs sampler with data augmentation algorithm is applied to the data from Taiwan Stock Market to estimate and test the model.
International Journal of Finance & Economics 6(3), pp.255-268