FORECASTING STOCK PRICE INDEX BY MULTIPLE REGRESSION

Over the last twenty years the financial markets of Hong Kong have developed rapidly. Although empirical studies on the behaviour of the Hong Kong stock market abound, much controversy over the efficiency of the market still exists. Some recent studies have shown that the market is inefficient in the “weak” form. Therefore one can justify employing the “fundamental approach” for stock price forecasting This study explores the use of multiple regression techniques to forecast stock price index. The results show that unemployment rate, trade balance, consumer price index and money supply are all significant in leading the stock price index. However, the regression models are still short of sufficient power to effectively predict change of direction of the index. Further enhancement of the models is needed if they are to be of real use for investment purposes.