Improving the accuracy: volatility modeling and forecasting using high-frequency data and the variational component

Manish Kumar


In this study, we predict the daily volatility of the S&P CNX NIFTY market index of India using the basic ‘heterogeneous autoregressive’ (HAR) and its variant. In doing so, we estimated several HAR and Log form of HAR models using different regressor. The different regressors were obtained by extracting the jump and continuous component and the threshold jump and continuous component from the realized volatility. We also tried to investigate whether dividing volatility into simple and threshold jumps and continuous variation yields a substantial improvement in volatility forecasting or not. The results provide the evidence that inclusion of realized bipower variance in the HAR models helps in predicting future volatility.


realized volatility, forecasting, time series analysis, autoregressive model

Full Text:



Licencia de Creative Commons 

This work is licensed under a Creative Commons Attribution 4.0 International License

Journal of Industrial Engineering and Management, 2008-2024

Online ISSN: 2013-0953; Print ISSN: 2013-8423; Online DL: B-28744-2008

Publisher: OmniaScience