Forecasting Stock Price Index Using Multiple Linear Regression Model

The stock market is one of the most popular methods for managing finances, and it has attracted more participants. Investing in stocks carries a comparatively significant risk. One of the most pressing issues in the stock market is how to lower risks while increasing returns for investors. This paper uses the clustering technique and multiple linear regression of machine learning. To analyze the different clustering techniques such as hierarchical, partitioning, and model-based technique validation index measures are used. K-means and model clustering algorithms have been tested, providing superior performance compared to hierarchical clustering techniques. Also, the accuracy of the multiple linear regression model is compared to time-series ARIMA models. Compared to ARIMA models, multiple regression analysis gives better accuracy and is used to predict future stock prices, which aids buyers and sellers in selecting firms from the stock.