Abstract

Stock price forecasting is the process of predicting future prices of company stocks using various analytical methods. This helps investors make informed decisions about buying or selling stocks. By analyzing past and current data, such as historical prices, trading volumes, and economic indicators, forecasters try to identify patterns and trends that indicate how stock prices might move in the future. Techniques used for this include statistical models. Time series forecasting is commonly used to predict future stock prices and analyze financial trends, which is crucial for guiding investors' decisions and trades. This project proposes an intelligent prediction system that uses sliding-window optimization to forecast stock prices with data science techniques. The system features a user-friendly interface and works as a standalone application. The proposed model is effective for predicting highly non-linear time series, which are challenging for traditional models. In this paper, we will use machine learning techniques such as ARIMA, Linear Regression, and Random Forest classifier to predict stock prices.