GAUSSIAN PROCESS FOR TURNING POINTS PREDICTION AND APPLICATION IN STOCK TRADING STRATEGY

Abstract

In the financial price sequence, the turning points prediction using the time series data form stock prices in Thai stock exchange. Turning points are critical local extreme points along a series. A trader who is able to buy stocks at trough prices and sell at peak prices to enter/exit the market precisely at the turning points would gain the maximum possible profit. In addition to using the Gaussian process model for predicting the turning points, in this project, we also test and compare the efficient techniques of modeling between the Gaussian Process Model, Neural Network and Support Vector Regression, respectively. Finally, we utilize the developed turning point prediction model to create trading strategy for the derivation of maximum profit.