No Detail Found!

Analysing trading strategies and forecasting stock prices using LSTM

Author(s)
RINKY DWIVEDI Maharaja Surajmal Institute of Technology,IDIA,
KOYEL DATTA GUPTA Maharaja Surajmal Institute of Technology,IDIA,
TRIPTI SHARMA Maharaja Surajmal Institute of Technology,IDIA,
RISHABH RAIZADA Maharaja Surajmal Institute of Technology,IDIA,
SARTHAK YADAV Maharaja Surajmal Institute of Technology,IDIA,
VEEPSA BHATIA Lyle School of Engineering Southern Methodist University, USA
Abstract
Algorithmic Stock Trading has been legalized in India since 2008, with SEBI (Securities and Exchange Board of India) regulating the norms governing the same. Earlier, regulations restricted third-party algorithms to be used in circulation with APIs (Application Programming Interface), as these were unregulated by registered brokerages. Owing to the COVID19 pandemic, SEBI relaxed these underlying norms, easing up both algorithm usage for an end-user, as well as the number of trade orders that can be placed per second (which went up from 20 to 120). This article aims to analyze optimal trading strategies for various stocks using both classic mathematical techniques and Recurrent Neural Networks (RNN). Stock prices will be forecasted using data analytics implementing indicators like supertrend and VWAP and machine learning models like LSTM. Upon fine-tuning parameters in both approaches, trend directions and triggers will be plotted (in the case of existing indicators and strategies) and predictions with trade triggers (in the case of LSTM). Furthermore, stocks will be categorized by trading techniques - growth, momentum, and value and further the best strategy with the expected profit percentage in each case will be identified.
Publication Details
Page(s) 4834-4843
DOI DOI not available
Published Journal: Journal of Theoretical and Applied Information Technology, Volume: 100, Issue: 15, Year: 2022

Request the full text from PASTIC

Keywords
forecasting, indicators, Stock Trading, Trading Strategies, LSTM
0

Citations

0

Downloads

35

Views