Generalised method of moments tests of mean-variance efficiency of the Australian equity market

2017-11-30T03:18:54Z (GMT) by Silvana Lau
For many years the Capital Asset Pricing Model (CAPM) developed by Sharpe (1964) and Lintner (1965) was the primary asset pricing model of financial theory. Over time, persistent criticism regarding the strict assumptions underlying the model resulted in numerous extensions of the model. Each extension involved relaxing one or more of the underlying assumptions. Unfortunately, empirical tests of these extensions have not proven to be ultimately superior. Early tests of the CAPM faced many problems ranging from fundamental conceptual problems to practical econometric problems. Some of these problems were overcome when Roll (1977) delivered his damaging critique about the CAPM. He basically concluded that the CAPM and all its extensions could never be tested because the true market index, the only regressor of the model, was not observable. Tests of the CAPM could thus only be interpreted as tests of whether or not the market index (the market's proxy) was mean-variance efficient. Tests of mean-variance efficiency tests whether or not a portfolio of risky assets has the greatest expected return for a given level of risk. In theory, the market portfolio should be a mean-variance efficient portfolio. [...]<br>