An empirical study on tactical asset allocation and forecasting

The central idea of this study is to analyse the moving average timing model that improves the risk-adjusted returns across various asset classes. This quantitative method tests Bombay Stock Exchange Index since 2000 on other diverse and publicly traded asset class indices, including the National Stock Exchange Index, Gold Index, Housing Development Finance Corporation Limited Mutual Fund Index and INR-Dollar Exchange Rates Index. This approach is then examined in a tactical asset allocation framework where the empirical results are equity-like returns with volatility, Sharpe ratio and drawdown. In this research, we also compared the forecasting performance of autoregressive moving average (ARMA) and exponential generalised autoregressive conditional heteroskedasticity-autoregressive moving average (EGARCH-ARMA) for the defined asset classes. Daily spot prices of all these composite indices provide the empirical sample for discussing and comparing the relative out-of-sample forecasting ability, given the growth potential of markets of India in the eyes of global investors. Empirical results indicate that the EGARCH-ARMA model is superior to the ARMA model in forecasting market returns. Several diagnostic tests were performed to select the models that best fit the index such as likelihood ratio test, Akaike and Bayesian information criteria tests and autocorrelation and partial autocorrelation tests.