Realized Tail-Risk Conditional Autoregressive Expectile Models: Accurate and Stable Forecasting Performance

By Keisuke Sasaki, Yasuhiro Omori

Rating

1850
Battle Count: 50

Relevance

9/10
Highly relevant for quantitative risk management, position sizing, and tail-risk hedging strategies. Accurate VaR and ES forecasts are critical for institutional trading desks and regulatory compliance.

Implementation Complexity

8/10
Requires implementation of MCMC algorithms (Metropolis-within-Gibbs), handling of high-frequency data to compute realized measures, and careful specification of priors and constraints. The use of pseudo-likelihood adds a layer of complexity compared to standard likelihood-based methods.

Reproducibility

3/5
The paper provides detailed mathematical specifications, prior distributions, and MCMC algorithms in the main text and supplementary material. The data source (Tick Data, LLC) is commercial and not publicly available, but the simulation design is fully specified. No public code repository URL is provided in the text.

Abstract

A realized tail-risk conditional autoregressive expectile (RTCARE) model is proposed for joint forecasting of value-at-risk (VaR) and expected shortfall (ES). Realized tail measures such as the realized quantile and the realized ES, constructed from intraday high-frequency returns, are incorporated into the dynamics and measurement equations of the conditional expectile, allowing information about lower-tail risk to enter the forecasts. The parameters are estimated by Markov chain Monte Carlo, and we evaluate the resulting VaR-ES forecasts in numerical examples and in an empirical application using U.S. stock-return data. Across the considered tail levels, the RTCARE models deliver competitive and often superior forecasting performance relative to existing alternatives, with stable results across Monte Carlo replications and rolling forecast evaluations.

Links

arXiv:2610.13446 ยท Paper rankings

The interactive Everscope explorer (charts, battles, favorites) loads below.