Conditional variance forecasts for long-term stock returns

• Verfasst von: Mammen, Enno [VerfasserIn]
• Verfasst von: Nielsen, Jens Perch [VerfasserIn]
• Verfasst von: Scholz, Michael [VerfasserIn]
• Verfasst von: Sperlich, Stefan [VerfasserIn]
• Titel: Conditional variance forecasts for long-term stock returns
• Verf.angabe: Enno Mammen, Jens Perch Nielsen, Michael Scholz and Stefan Sperlich
• Jahr: 2019
• Umfang: 22 S.
• Titel Quelle: Enthalten in: Risks
• Ort Quelle: Basel : MDPI, 2013
• Jahr Quelle: 2019
• Band/Heft Quelle: 7(2019), 4/113 vom: Dez., Seite 1-22
• ISSN Quelle: 2227-9091
• DOI: doi:10.3390/risks7040113
• Datenträger: Online-Ressource
• Sprache: eng
• Sach-SW: autocorrelation
• Sach-SW: benchmark
• Sach-SW: cross-validation
• Sach-SW: long-term forecasts
• Sach-SW: overlapping returns
• Sach-SW: prediction
• Sach-SW: stock return volatility
• Form-SW: Aufsatz in Zeitschrift
• K10plus-PPN: 1684080657

Abstract

In this paper, we apply machine learning to forecast the conditional variance of long-term stock returns measured in excess of different benchmarks, considering the short- and long-term interest rate, the earnings-by-price ratio, and the inflation rate. In particular, we apply in a two-step procedure a fully nonparametric local-linear smoother and choose the set of covariates as well as the smoothing parameters via cross-validation. We find that volatility forecastability is much less important at longer horizons regardless of the chosen model and that the homoscedastic historical average of the squared return prediction errors gives an adequate approximation of the unobserved realised conditional variance for both the one-year and five-year horizon.