Best-Arm Identification Using Extreme Value Theory Estimates of the CVaR

Journal of Risk and Financial Management

Dylan Troop¹  Jia Yuan Yu¹  Frédéric Godin² 

[1] Concordia Institute of Information System Engineering, Concordia University, Montréal, QC H3G 1M8, Canada;Department of Mathematics and Statistics, Concordia University, Montréal, QC H3G 1M8, Canada;

关键词: sequential decision making;    multi-armed bandits;    conditional value-at-risk;    extreme value theory;    heavy-tailed distributions;    risk-aware reinforcement learning;   

DOI  :  10.3390/jrfm15040172

来源: DOAJ

【 摘 要 】

We consider a risk-aware multi-armed bandit framework with the goal of avoiding catastrophic risk. Such a framework has multiple applications in financial risk management. We introduce a new conditional value-at-risk (CVaR) estimation procedure combining extreme value theory with automated threshold selection by ordered goodness-of-fit tests, and we apply this procedure to a pure exploration best-arm identification problem under a fixed budget. We empirically compare our results with the commonly used sample average estimator of the CVaR, and we show a significant performance improvement when the underlying arm distributions are heavy-tailed.

【 授权许可】

Unknown