# Forecasting NIFTY 50 benchmark Index using Seasonal ARIMA time series models

**Authors:** Amit Tewari  
**arXiv:** [2001.08979](https://arxiv.org/abs/2001.08979)  
**Date:** 2020-01-09  
**Summary page:** https://www.summarizepaper.com/en/paper/2001.08979v1/

## Key points

- Time Series Analysis techniques applied to Nifty 50 index
- Focus on using Seasonal Auto Regressive Integrated Moving Average (SARIMA) models for forecasting
- Evaluation of 729 model parameter combinations
- Selection of most appropriate model based on Akaike Information Criterion (AIC)
- Nifty 50 index represents weighted average of 50 Indian company stocks across sectors
- Used for benchmarking fund portfolios, launching index funds, ETFs, and structured products
- Reflects overall Indian stock market and holds significant importance
- Introduction to ARIMA models used in analysis
- ARIMA models explain time series based on past values and forecast future values
- Consist of three terms: p (autoregressive term), d (differencing for stationarity), q (moving average term)
- SARIMA models capture patterns and predict future movements in non-seasonal time series data like Nifty 50
- Research contributes to understanding how Time Series Analysis can be used for forecasting stock market indices like Nifty 50 with implications for investors, fund managers, and financial institutions.

## Summary

This paper analyzes the application of Time Series Analysis techniques to capture the movement of the Nifty 50 index, which is one of the most actively traded contracts globally in the stock market. The study specifically focuses on using Seasonal Auto Regressive Integrated Moving Average (SARIMA) models to forecast the behavior of this benchmark index. A total of 729 model parameter combinations were evaluated, and the most appropriate model was selected based on Akaike Information Criterion (AIC) criteria. The Nifty 50 index, operated by the National Stock Exchange of India, represents a weighted average of 50 Indian company stocks across various sectors. It is considered one of the main stock indices in India alongside BSE Sensex and is widely used for purposes such as benchmarking fund portfolios, launching index funds, exchange-traded funds (ETFs), and structured products. As it tracks the behavior of blue-chip companies and reflects the overall Indian stock market, it holds significant importance. The paper also provides an introduction to ARIMA models which are employed in this analysis. ARIMA models explain a given time series based on its own past values and can be used to forecast future values. These models consist of three terms: p (order of autoregressive term), d (number of differencing required for stationarity), and q (order of moving average term). By applying SARIMA models to non-seasonal time series data like that of Nifty 50, patterns can be captured and future movements can be predicted. Overall, this research contributes to understanding how Time Series Analysis techniques can be utilized for forecasting stock market indices like Nifty 50 with implications for investors, fund managers, and financial institutions who rely on accurate predictions for decision-making purposes.