Modelling Interdependent Returns and Risk in the Environmental Industry

This paper stresses the importance of assessing the risk-return trade-off faced by environmental industries in financial markets. One of the most widely-used theoretical models in finance is the conditional CAPM, which describes the conditional risk-return tradeoff in financial markets, whereby both the conditional mean return and conditional beta risk are allowed to vary over time. This paper models the time-varying conditional mean and variances of returns, and the correlations between the individual asset return and the market return in the conditional CAPM in a multivariate GARCH framework. Specifically, the Vector ARMAAsymmetric Multivariate GARCH model of Chan, Hoti and McAleer (2003), which incorporates the CCCMGARCH of Bollerslev (1990) and the Vector ARMA-GARCH of Ling and McAleer (2002), is applied to the conditional CAPM, with the conditional CAPM specification replacing the ARMA specification in the conditional mean equation. The main motivation for using the Conditional CAPM-AMGARCH-in-mean model specification is not only because it has the ability to capture the stylized facts of financial asset returns such as persistence of volatility, volatility clusters and excess kurtosis, but it also considers the interdependencies of volatilities between the individual asset return and the market return, and the asymmetric effects of the unconditional shocks on the conditional variances of the individual asset return and the market return respectively. Further, the structural and statistical properties of this model have been established in Chan et al. (2003). Our dataset consists of monthly excess returns on the Australian Mining industry sectors including Gold Mining, Other Mining and Mining Finance from the period January 1980 to December 2002. The paper suggests that the conditional CAPM is inadequate in explaining the financial risk-return tradeoff for the environmental industry sectors; however, there appears to be some interdependent ARCH/GARCH effects between the Other Mining and Mining Finance excess returns and the market excess return, and no asymmetric ARCH/GARCH effects.