This paper employs a multifactor model motivated by the Arbitrage Pricing Theory (APT) to describe the time series behaviour of the South African stock market as represented by the JSE All-Share Index. Factors representative of eight risk factor categories are considered, namely inflation, real activity, the money supply, interest rates, commodities, exchange rates, business cycles indicators and international market indices. These categories are then represented in the return generating process. The results indicate that the South African stock market is influenced by movements in international markets, inflation, inflation expectations, real activity, the money supply, oil prices, exchange rates and cyclical variations in the business cycle. Furthermore, the Autoregressive Conditional Heteroscedastic (ARCH) and Generalized Autoregressive Conditional Heteroscedastic (GARCH) model framework is found to be a more appropriate econometric framework relative to the Least Squares framework (LS) for models of the return generating process of South African stock returns.
|Journal||Journal for Studies in Economics and Econometrics|
|Publication status||Published - Apr 2015|