Stock Market Prediction with Lasso Regression using Technical Analysis and Time Lag

Stock price prophecy is one of the most recent topics of exploration in both university and business. Stock market is based on psychology of traders and it's uncertain but this uncertainty can be dodged based upon technical analysis. Though real live trading data have a few features but based upon the calculations of these less features more features are created that makes prediction an organized task. In this paper Least Absolute Shrinkage and Selection Operator (LASSO) Regression along with added features based on Technical Analysis such as Moving Averages, Relative Strength Index, Super Trend and Time Lag calculations, a proposed novel method to predict stock prices. The model is able to perform very well in terms of prediction on features that are added considering further days prices as targets. National Stock Exchange (NSE) NIFTY 50 index stock data is taken for empirical calculations.

I. Introduction

Stock price prophecy has been one of the latest research topics in recent years. It is the act of trying to predict the future value of a company stock. The successful prediction of a stock's future price might perform a significant profit. Predicting the future of stock market move is one of the most challenging issues because of many factors involved in the stock market, such as company fundamentals, politics, public demands, interest rates, and economic growth that make the stock market volatile and difficult to predict correctly. The prediction of shares offers big chances for profit and is a main motivation for research in this area. Stock market data contains many attributes, far more than trader can comprehend. Prediction methodologies falls into two broad categories which can often do overlap. Fundamental analysis and technical analysis are the two types of stock analysis.