A Comprehensive Review of AI-Driven Techniques for Predicting Stock Market Trends

Stock market prediction is essential in providing investors and companies with data-driven financial decisions. An accurate prediction of stock prices significantly enhances profit potential while reducing investment risks, making it invaluable in modern finance. This paper explores the application of Linear Regression, widely adopted and fundamental machine learning algorithm, for predicting stock prices. Linear Regression assumes that there is a linear relationship of the stock price (response variable) with several variables that could be independent like trading volume, historical prices, and other market-based indicators. The algorithmic approach seeks to fit a line of best fit which minimizes the discrepancy between the calculated and actual stock prices giving a model that is as simple as it is computationally efficient and interpret-able. Linear Regression provides easy clearances on how each factor contributes towards price movement. However, in complex patterns or with non-linear trends that might prevail in financial data sets, the performance of a model may be limited due to these features. Authors in this paper will have aimed at enhancing the prediction of the model by including several feature engineering techniques applied on the input variables followed by their refinement. More so, we outline how Linear Regression may be combined with other more advanced techniques, such as moving averages or sentiment analysis, to fine-tune the outcome even further.