Comparative analysis of different machine learning models for predicting Google stock price

Abstract

With an increasing reliance on data-driven decision-making in the business world, the need for accurate stock price predictions has grown substantially. This paper aims to address this crucial aspect, aligning machine learning methodologies with real-world business applications. The dynamic nature of financial markets indicates that it is characterized by intricate price movements influenced by an array of factors. Because it has the ability to reveal patterns and relationships in the data that traditional research can miss, machine learning has attracted a lot of attention in this context when it comes to predicting stock values. Methodologically, this paper applies three popular machine learning models to predict Google's stock prices. Linear Regression, Decision Tree, and Random Forest are each employed to assess their predictive accuracy and reliability. Metrics such as Root Mean Squared Error (RMSE), R-squared, Mean Absolute Error (MAE), and Residual Analysis are used to evaluate the models' performance. The findings reveal that Linear Regression outperformed the other models in predicting Google's stock prices. Its superior performance, as indicated by the evaluation metrics, suggests that it may be a valuable tool in the realm of financial forecasting. These findings highlight how data-driven insights can improve company strategy and decision-making, and they have important ramifications for the use of machine learning in stock market analysis.