Forecasting Realised Volatility: Does the LASSO approach outperform HAR?

39 Pages Posted: 22 Mar 2021

See all articles by Yi Ding

Yi Ding

University of Southampton - Department of Banking and Finance

Dimos S Kambouroudis

University of Stirling - Department of Accounting and Finance

David G. McMillan

University of Stirling

Date Written: March 11, 2021

Abstract

The HAR model dominates current volatility forecasting. This model implies a restricted lag approach, with three parameters accounting for an AR(22) structure. This paper uses the Lasso method, which selects a parsimonious lag structure, while allowing both a flexible lag structure and lags greater than 22. In-sample results suggest that while significance is largely found among the first 22 lags, consistent with the HAR model, there is evidence that longer lags contain information, as Lasso models providing an improved fit. Out-of-sample forecasts for daily, weekly and monthly volatility, evaluated using MSE, QLIKE, MCS and VaR measures suggest that the ordered Lasso model provides the preferred forecasts using an AR(100) at the daily level and an AR(22) for the weekly and monthly horizons. The results support the view that a more flexible lag structure is preferred over the HAR approach.

Keywords: Volatility Forecasting, HAR, Lasso, VaR

JEL Classification: C22, G12

Suggested Citation

Ding, Yi and Kambouroudis, Dimos S and McMillan, David G., Forecasting Realised Volatility: Does the LASSO approach outperform HAR? (March 11, 2021). Available at SSRN: https://ssrn.com/abstract=3802466 or http://dx.doi.org/10.2139/ssrn.3802466

Yi Ding

University of Southampton - Department of Banking and Finance ( email )

Southampton
United Kingdom

Dimos S Kambouroudis

University of Stirling - Department of Accounting and Finance ( email )

Stirling, Scotland FK9 4LA
United Kingdom

David G. McMillan (Contact Author)

University of Stirling ( email )

Stirling, Scotland FK9 4LA
United Kingdom

Do you have negative results from your research you’d like to share?

Paper statistics

Downloads
148
Abstract Views
604
Rank
356,750
PlumX Metrics