Predicting future Stock market Trends using Python and Machine Learning

Investors have long grappled with the ambiguous nature of the stock market movements due to a multitude of influential factors, one of the factors is trying to predict the changes in the trends of the stock market. To address this challenge, the objective of this study is to substantially mitigate the risk associated with trend prediction through the utilization of machine learning algorithms. In this study, a comparison of five machine learning models: Support Vector Machines (SVM), K-Nearest Neighbors (KNN), Linear Regression, Lasso Regression, and Bayesian Regression and each prediction model are assessed using two metrics, namely R-square and mean absolute error, which is based on different input approaches. The input values consist of ten years of historical data, and two methods are considered for their application. Firstly, the indicators are calculated by utilizing continuous data derived from stock trading values. Secondly, features are extracted from both dependent and independent variables. Additionally, results have shown that SVM offers good accuracy and performs faster prediction compared to other models like KNN, Linear Regression, Lasso Regression, and Bayesian Regression.