Comparative Study among ARIMA, SARIMA & XGBoost for Prediction of NIFTY IT Index

No Thumbnail Available
Authors
Bendale, Mayuri
Issue Date
2024
Degree
MSc in Business Analytics
Publisher
Dublin Business School
Rights
Abstract
Nifty IT index of India stock market is one of the most important yet neglected index when it comes to research and prediction. Prediction of Nifty IT index has benefits would be able to provide foresight and informed decision making to investors, traders, policy makers, etc. as Nifty IT represents IT sector of India. This research is a comparative study between three time series prediction algorithms viz. ARIMA, SARIMA and XGBoost for the most precise forecasting of Nifty IT index. The dataset used for this study has the Nifty IT index data of last 6 years. This time frame covers the dramatic historic moments such as covid-19 pandemic, Russia-Ukraine war, India-Canada tensions and the drastic changes in prices during these events. Three models were hyper parameter tuned and then compared on the basis of three metrices- MSE, RMSE and MAE. Out of the three, SARIMA models seem to have outperformed both ARIMA and XGBoost and hence the conclusion of the study is SARIMA is the most precise algorithm to use for prediction of Nifty IT index out of three.