Abstract

Stock market prediction is a key problem to the financial field. An accurate prediction model may yield profits for investors. However, the volatile nature of the stock market makes it difficult to develop efficient models. This study attempts to improve the prediction capacity of the stock price via an integrated prediction model based on Kernel Principal Component Analysis (KPCA) and Support Vector Machines for Regression (SVR). KPCA is firstly introduced to reduce the feature dimensions. On that basis, SVR is used to build a short-term investment decision system. 23 technical indicators are calculated for five major Moroccan banks listed on the Casablanca Stock Exchange (Casablanca S.E) and used as input of the models. A comprehensive parameters setting is performed to improve the prediction performance. The simulation results show that, through KPCA attribute reduction, the structure of the investment decision system can be simplified significantly with improvement of the model performance. The average performance of the integrated model that uses KPCA and SVR is significantly better than that of SVR model, which verifies the effectiveness and accuracy of the proposed method.

Full Text
Published version (Free)

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call