South African stock return predictability in the context data mining : the role of financial variables and international stock returns

摘要

In this paper, we examine the predictive ability, both in-sample and the out-of-sample,for South African stock returns using a number of financial variables, based on monthlydata with an in-sample period covering 1990:01 to 1996:12 and the out-of-sample periodof 1997:01 to 2010:04. We use the t-statistic corresponding to the slope coefficient in apredictive regression model for in-sample predictions, while for the out-of-sample, theMSE-F and the ENC-NEW tests statistics with good power properties were utilised. Toguard against data mining, a bootstrap procedure was employed for calculating the criticalvalues of both the in-sample and out-of-sample test statistics. Furthermore, we use aprocedure that combines in-sample general-to-specific model selection with out-ofsampletests of predictive ability to further analyse the predictive power of each financialvariable. Our results show that, for the in-sample test statistic, only the stock returns forour major trading partners have predictive power at certain short and long run horizons.For the out-of-sample tests, the Treasury bill rate and the term spread together with thestock returns for our major trading partners show predictive power both at short andlong run horizons. When accounting for data mining, the maximal out-of-sample teststatistics become insignificant from 6-months onward suggesting that the evidence of theout-of-sample predictability at longer horizons is due to data mining. The general-tospecificmodel shows that valuation ratios contain very useful information that explainsthe behaviour of stock returns, despite their inability to predict stock return at anyhorizon. The model also highlights the role of multiple variables in predicting stockreturns at medium- to long-run horizons.