The prediction of a particular stock price serves as recommendation system for investors. Most of stock prediction studies focus on using macroeconomic indicators to train the prediction model. Due to difficulty in obtaining this data on daily basis, we directly employ the daily prices data to train the model for predicting the stock price. This study focuses on identifying significant inputs among the financial indicators using Principal Component Analysis to construct a model for prediction. A Multilayer Feed-Forward Nonlinear Autoregressive with External (Exogenous) Input (NARX) network is trained and used to predict closing price of a share listed in National Stock Exchange (NSE). Financial indicators of State Bank of India (SBI) are used as case study to train & test the proposed system. NARX network designed for year 2012 and tested for year 2013.

Artificial Neural Network (ANN), Nonlinear Autoregressive with External Input (NARX), Principal Component Analysis (PCA), stock price prediction

[1] Kao, L-J., Chiu, C-C., Lu, C-J. and Yang, J-L. (2013) ‘Integration of nonlinear independent component analysis and support vector regression for stock price forecasting’, NeuroComputing, Vol. 99, No. 1, pp.534–542
[2] Wang, Y. (2014) ‘Stock price direction prediction by directly using prices data: an empirical study on the KOSPI and HSI’, Int. J. Business Intelligence and Data Mining, Vol. 9, No. 2, pp.145–160.
[3] J. Edward Jackson, `A User`s Guide To Principal Components`, A Wiley-Interscience Publication, Page 1
[4] Abhyankar, A. Copeland, L.S., & Wong, W., Uncovering nonlinear structure in real-time stock-market indexes: The S&P 500, the DAX, the Nikkei 225, and the FTSE-100, Journal of Business & Economic Statistics, 15, 1–14., 1997
[5] Mbeledogu N. N., Odoh M. And Umeh M.N., ‘Stock Feature Extraction using Principal Component Analysis’, 2012 International Conference on Computer Technology and Science, IPCSIT vol. 47
[6] Marijana Zekić-Sušac, Nataša Sarlija, and Sania Pfeifer, "Combining PCA Analysis and Artificial Neural Networks in Modelling Entrepreneurial Intentions of Students", Croatian Operational Research Review (CRORR), Vol. 4, 2013
[7] Richard A. Johnson, Dean W. Wichern, `Applied Multivariate Statistical Analysis`, 6th Edition, Pearson Prentice Hall, Page 430
[8] Tsungnan Lin, Bill G. Horne, Peter Tino, C. Lee Giles, Learning long-term dependencies in NARX recurrent neural networks, IEEE Transactions on Neural Networks, Vol. 7, No. 6, 1996, pp. 1329-1351
[9] Yang Gao, Meng Joo Er, NARMAX time series model prediction: feed-forward and recurrent fuzzy neural network approaches, Fuzzy Sets and Systems, Vol. 150, No. 2, 2005, pp.331-350
[10] Prashant S Chavan, Prof. Dr. Shrishail. T. Patil, ‘Parameters for Stock Market Prediction’, Int. J. Computer Technology & Applications, Vol 4 (2), 337-340 , ISSN:2229-6093
[11] www.nseindia.com/products/content/equities/eq_security.htm
[12] Jiawei Han, Micheline Kamber, Jian Pei, Data Mining Concepts and Techniques, MK, 3rd edition, Page 113-114
[13] Support.sas.com/documentation/cdl/en/statug/67523/HTML/default/viewer.htm#statug_princomp_overview.htm
[14] https://en.wikipedia.org/wiki/SAS_(software)