CS134 Project Proposal

A Hybrid Machine Learning System for Stock Market Forecasting

My email: Jianfeng.Huo@dartmouth.edu


Motivation:

Recently, a large amount of amazing work has been done in the area of analyzing and predicting stock prices and index changes using Machine Learning Algorithms. Intelligent Trading Systems has been used for most of the stock traders to help them in predicting prices based on various situations and conditions, thereby helping them in making instantaneous investment decisions. Stock market prediction is regarded as one of the most challenging task in financial time-series forecasting. This is primarily because the underlying nature of the uncertain financial domain and in part because of the mix of known parameters (Previous Day’s Closing Price, P/E Ratio etc.) and unknown factors (Election Results, Rumors etc.).

 

In my project, I will implement a hybrid machine learning system which is proposed by Rohit Choudhry, and Kumkum Garg [1] based on Genetic Algorithm (GA) and Support Vector Machines (SVM) for stock market prediction. A variety of indicators from the technical analysis field of study are used as input features. We also make use of the correlation between stock prices of different companies to forecast the price of a stock, making use of technical indicators of highly correlated stocks, not only the stock to be predicted. The genetic algorithm is used to select the set of most informative input features from among all the technical indicators. The results show that the hybrid GA-SVM system outperforms the stand alone SVM system. The data used for this study were obtained from the Yahoo Finance website [2].

 

Before milestone, I expect to implement the hybrid machine learning system in Matlab. After milestone, I will focus on the comparison the experience results and to add new features to make it more efficient and accurate.

 


Timeline:

l  Read papers about using GA and SVM for stock market prediction

l  Before milestone implement the hybrid machine learning system in Matlab.

l  Before final debug the algorithm, test data and improve accuracy of the system.

l  Analyze experiment results and write the final report.


Reference

[1] R. Choudhry and K. Garg, "A hybrid machine learning system for stock market forecasting," Proceedings of World Academy of Science, Engineering and Technology, vol.29, pp. 315-318, 2008. 

[2] http://in.finance.yahoo.com/

 


Other Resources

[1] L. J. Cao and F. E. H. Tay, "Financial forecasting using support vector machines", Neural Comput. Applicat., vol. 10,  no. 2,  pp. 184-192, 2001.  

[2] Vatsal H. Shah, “Machine Learning Techniques for Stock Prediction”, http://www.vatsals.com/Essays/MachineLearningTechniquesforStockPrediction.pdf