A Survey on Machine Learning Techniques for Stock Market Price Prediction

In today's fiercely competitive business environment, industries aspire to grow rapidly. Achieving consistent business expansion requires access to substantial capital resources. Typically, there are three primary avenues for raising capital: Initial Public Offering (IPO), securing investments from angel investors, and obtaining business loans. As companies grow beyond a certain size, it becomes increasingly challenging for individual investors to sustain operations solely from their capital. Consequently, businesses require a continuous influx of capital. An IPO is a prominent method for raising capital, known to elevate a company's profile. It bolsters the company's credibility among suppliers, collaborators, and customers. While the primary motivation for an IPO is capital infusion, it may also offer founders and early-stage investors an opportunity for partial divestment of their holdings. Following an IPO, a company's shares become publicly tradable. It offers individuals the opportunity to invest in the form of stocks traded on open platforms. However, achieving optimal returns on investments necessitates the ability to forecast stock prices accurately. This survey explores various techniques for the prediction of stock prices, all rooted in domain of Machine Learning (ML). A comprehensive review of existing research indicates that combining Random Forest with other models may represent the most effective approach to stock price prediction. It also proposes potential avenues for enhancing future research in this domain.