Predicting Direction Of Stock Price Index Movement Based On Support Vector Machines

Predictions of stock market price and its direction are quite difficult. In re-sponse to such difficulty, machine learning techniques have been introduced and applied for this financial predict ion. There exists vast research articles which predict the stock market but most of the proposed models focus on the accu-rate forecasting of the levels of the underlying stock index. There is a lack of studies examining the predictability of the direction of stock index movemen-t,especially in the emerging market,such as China. Given the notion that a prediction of the direction is a practical issue which usually affects a financial trader’s decision to buy or sell an instrument and to make effective market trading strategics.This study uses Support Vector Machinc(SVM)to predict the daily move-ment direction of China Security Index 300(CSI300).Ten technical indicators were selected as input variables.The directions of daily change of China Secu-rity Index 300 arc categorized as "0" and "1". "0" means that China Secu-rity Index 300 at time t is lower than that at time t — 1, while China Security Index 300 at time t is higher than that at time t — 1, the direction is "1".In order to get a better outcome,We also use the Vector Autoregression(VAR)to predict the stock index price.Finally,We find that combining the VAR and SVM can improve the prediction accuracy.