Lasso-based forecast combinations for forecasting realized variances

TL;DR

This paper explores the use of various lasso-based methods, including extensions like ordered lasso, for forecasting realized variances in stock markets, demonstrating improved accuracy especially with multivariate models over longer horizons.

Contribution

It introduces and evaluates lasso extensions for volatility forecasting, highlighting the effectiveness of ordered lasso and multivariate models in financial applications.

Findings

Ordered lasso yields the most accurate forecasts.

Multivariate models outperform univariate models for longer horizons.

Lasso extensions effectively handle model parsimony and dynamic features.

Abstract

Volatility forecasts are key inputs in financial analysis. While lasso based forecasts have shown to perform well in many applications, their use to obtain volatility forecasts has not yet received much attention in the literature. Lasso estimators produce parsimonious forecast models. Our forecast combination approach hedges against the risk of selecting a wrong degree of model parsimony. Apart from the standard lasso, we consider several lasso extensions that account for the dynamic nature of the forecast model. We apply forecast combined lasso estimators in a comprehensive forecasting exercise using realized variance time series of ten major international stock market indices. We find the lasso extended "ordered lasso" to give the most accurate realized variance forecasts. Multivariate forecast models, accounting for volatility spillovers between different stock markets, outperform univariate forecast models for longer forecast horizons.