Harnessing Machine Learning and Deep Learning to Anticipate Stock Price Variations Post-New Product Launches

There are many algorithms that work behind stock market analysis and hence stock market trading has turned out towards the greatest and most existing activity when one speaks about the financial markets. Uncertainty and volatility in the prices of the stocks keep on making the investor look for means to predict future trends in order to avoid losses and get the maximum possible profits; however, one cannot deny that, up till now, no such technique can predict the upcoming trends in the markets with 100 percent accuracy, yet various methods are being experimented for improving the predictive performance of models to a higher extent. In the past few years, there are numerous algorithms deployed for stock price prediction because of the rapid progress of machine learning and deep learning. The research has concentrated on five algorithms: K-nearest neighbors, linear regression, support vector regression, decision tree regression, and long short-term memory to predict the stock prices of $\mathbf{1 2}$ leading Indian stock market companies. With extensive background research done in and around an attempt to apply Machine Learning in the stock markets, this particular study has been indeed an exhaustive one-leveraging precise and quantitative data of 12 mentioned companies acquired through 7 years of experience. In addition, other efficient and robust approaches which are used to forecast market trends are discussed. At the same time, a detailed stepwise methodology for obtaining the results is discussed in this paper. In addition to that, a comparative analysis of the performances of the algorithms mentioned for stock price prediction has been performed showing the results in a tabulated and graphical form so that it is easier to comprehend.