The wide adoption of machine learning techniques in predicting stock prices has led to the emergence of many articles on the topic. Howev-er, a systematic review on the topic remains lacking. This paper provides a systematic review of the recent applications of machine learning techniques in the construction of stock prediction models. A framework is designed to classify and evaluate the relevant work in recent arti-cles based on the type of model, type of financial market, type of prediction technique, type of optimization approach, type of indicators, type of performance metrics, type of benchmark models and prediction results. It is observed that financial indicators are the frequently used input variables and different forms of machine learning techniques are integrated to predict the stock prices. There are 4 variables that im-pose significant influence on the prediction model, namely the type of input variables, type of prediction technique, type of optimization approach and number of analysis layer. Thus, the limitations and potential enhancement on the 4 variables are discussed so that optimal combinations will be established in future research efforts.
Tan et al. (Fri,) studied this question.
Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context: