A Statistical Regression Analysis of Financial Time Series Using ARIMA Models

To take a sound strategic investment decision, prediction of financial markets becomes indispensable. The time series forecasting finds application in various traditional domains like the Auto-Regressive Integrated Moving Average (ARIMA) model as the model fully leverages its underlying capability of capturing trends and patterns in the historical data. This paper discusses the ARIMA model for forecasting life via torrent in the financial market and employs stock market index values from the Bombay Stock Exchange (BSE). The study revealed that ARIMA works particularly well for short-term forecasting, which gives essential market trend indications. This proves its usefulness for financial forecasting and also offers some valuable observations to investors and analysts wishing to study market behavior and improve forecasting accuracy.