Predicting trends in the financial markets remains a demanding yet crucial responsibility undertaken by market participants and financial professionals. This research quantitatively evaluates how effectively machine learning algorithms anticipate outcomes in stock price forecasting for five major Indian companies: Reliance Power, Hindustan Unilever, State Bank of India, Tata Motors, and Nestle. By integrating traditional statistical techniques, including simple moving average (SMA) and exponential moving average (EMA), with supervised learning algorithms such as logistic regression, k-nearest neighbors (KNN), and linear discriminant analysis (LDA), this study provides a comparative analysis of model effectiveness. The empirical findings reveal that classification-based models exhibit varying degrees of predictive efficacy. LDA correctly classified 65% of original grouped cases, surpassing the minimum required accuracy of 63. 37%. KNN achieved a higher accuracy of 75%, outperforming LDA. Logistic regression demonstrated an improvement from the intercept model’s 55. 9% accuracy to 57. 1% in Block 1 model, indicating that incorporating meaningful variables enhances prediction reliability. Additionally, EMA proved to be more responsive to price volatility, effectively capturing short-term fluctuations compared to SMA. The results provide meaningful understanding into the strengths (Sayan Hazra, 2023, 10 April, Implementation and performance evaluation of standard multi-class classification algorithms using mnist data set, Medium, https: //medium. com/@sayanrik1996₃4278/implementation-and-performance-evaluation-of-standard-multi-class-classification-algorithms-using-8d2dc917143) and limitations of different machine learning algorithms for stock market prediction. Future research can explore hybrid modeling approaches and deep learning techniques to further refine predictive accuracy and enhance algorithmic trading strategies. JEL Classification: C22, C45, C53, G11, G17
Karulkar et al. (Wed,) studied this question.