Importance of Machine Learning in Making Investment Decision in Stock Market

Abstract

Executive Summary Predicting stock trends in the financial market is always demanding but satisfying as well. With the growing power of computing and the recent development of graphics processing unit and tensor processing unit, analysts and researchers are applying advanced techniques such as machine learning techniques more and more to predict stock price trends. In recent years, researchers have developed several algorithms to predict stock trends. To assist investors interested in investing in the stock market, preferably for a short period, it has become necessary to review research papers dealing on machine learning and analyse the importance of their findings in the context of how stock price trends generate trading signals. In this article, to achieve the stated task, authors scrutinized more than 50 research papers focusing on various machine learning algorithms with varied levels of input variables and found that though the performance of models measured by root-mean-square error (RMSE) for regression and accuracy score for classification models varied greatly, long short-term memory (LSTM) model displayed higher accuracy amongst the machine and deep learning models reviewed. However, reinforcement learning algorithm performance measured by profitability and Sharpe ratio outperformed all. In general, traders can maximize their profits by using machine learning instead of using technical analysis. Technical analysis is very easy to implement, but the profit based on it can vanish too soon or making a profit using technical analysis is almost difficult because of its simplicity. Hence, studying machine, deep and reinforcement learning algorithms is vital for traders and investors. These findings were based on the literature review consolidated in the result section.

Cite this work

No public copy is stored for this record, so only the metadata is cited.

Stored only when both papers are in the corpus — 35 citations exist upstream.

Papers that cite the same references — the relational fallback while embeddings are still backfilling.