Survey of Stock Market Price Prediction Trends using Machine Learning Techniques

Abstract

Investing in the stock market is an essential aspect of the financial sector. However, the task of identifying lucrative stocks is a challenging one that requires careful analysis. This study aims to address this challenge by comparing various Machine Learning and Deep Learning techniques for predicting stock trends. The research evaluates and compares different models, including Long Short-Term Memory (LSTM), Prophet (Automated Forecasting Procedure), Random Decision Forest, Auto-ARIMA, k-Nearest Neighbors (KNN), Linear Regression, and Moving Average techniques like SMA and EMA. Furthermore, a new hybrid model is proposed, which outperforms existing models in terms of accuracy. The models are trained and tested on a historical dataset of stocks from different industrial sectors and evaluated based on various performance metrics. The study provides insights into the accuracy of different prediction models and can help investors, traders, and financial analysts make informed investment decisions. Additionally, the findings of this research work can serve as a benchmark for future research on stock market prediction.