A0428
Title: HARd to beat: The overlooked impact of rolling windows in the era of machine learning
Authors: Jonathan Chassot - University of St.Gallen (Switzerland) [presenting]
Francesco Audrino - University of St Gallen (Switzerland)
Abstract: The predictive abilities of the heterogeneous autoregressive (HAR) model compared to machine learning (ML) techniques are investigated across an unprecedented dataset of 1'445 stocks. The analysis focuses on the role of fitting schemes, particularly the training window and re-estimation frequency, in determining the HAR model's performance. Despite extensive hyperparameter tuning, ML models fail to surpass the linear benchmark set by HAR when utilizing a refined fitting approach for the latter. Moreover, the simplicity of HAR allows for an interpretable model with drastically lower computational costs. Performance is assessed using QLIKE, MSE, and realized utility metrics, finding that HAR consistently outperforms its ML counterparts when both rely solely on realized volatility and VIX as predictors. The results underscore the importance of a correctly specified fitting scheme. They suggest properly fitted HAR models provide superior forecasting accuracy, establishing robust guidelines for their practical application and use as a benchmark. The efficacy of the HAR model is not only reaffirmed but also a critical perspective on the practical limitations of ML approaches is provided in realized volatility forecasting.
