Forecasting Sectoral Indices in The Kuala Lumpur Stock Exchange

Abstract

This paper evaluates the use of (i) Box-Jenkins autoregressive-moving average model, (ii) vector model that incorporates short-run intersectoral relationship, and (iii) vector error model that incorporates long-run intersectoral relationship, for forecasting the daily Finance, Plantation, Mining and Property Index of the Kuala Lumpur Stock Exchange. Given its explanation the behaviour of the stock prices, the random walk was used as a benchmark. of the long-run equilibrium sectoral relationship was found to track rather closely the one- forecasts of a random walk. The autoregressive-moving average model follows next, and the regression model has the poorest performance.