0598425 20250317085712.320240921235959.9 001323540900024 10.1007/978-3-031-65993-5_24 8 s. P cav_un_epca*0598452978-3-031-65992-82194-5357200-207ChamSpringer2024 conditional-value-at-risk averaged regression quantile two-step regression quantile cav_un_auth*0368969 Jurečková Jana UTIA-B Stochastická informatika Department of Stochastic Informatics SI SI CZ 33,3 A Ústav teorie informace a automatizace AV ČR, v. v. i. cav_un_auth*0019770 Picek J. CZ cav_un_auth*0263018 Kalina Jan UIVT-O Oddělení umělé inteligence Department of Artificial Intelligence Department of Machine Learning Ústav informatiky AV ČR, v. v. i. https://library.utia.cas.cz/separaty/2024/SI/jureckova-0598425.pdf https://link.springer.com/chapter/10.1007/978-3-031-65993-5_24 GA22-03636S GA ČR CZ cav_un_auth*0435411 GA24-11146S GA ČR CZ cav_un_auth*0474719 The conditional value-at-risk (CVaR) represents a popular risk measure often exploited e.g. within portfolio optimization. The situation with a nuisance linear regression is considered here, in other words, we do not observe directly the loss Z of interest, but only Y=\beta _0+X\beta+Z, where the covariates are not under our control. We propose a novel estimator of CVaR(Z) based on the averaged two-step regression quantile combined with an R-estimate of regression parameters. cav_un_auth*0472961 International Conference on Soft Methods in Probability and Statistics 2024 - SMPS 2024 /11./ 20240903 20240906 Salzburg AT BB 10000 10100 10103 2025 3 UIVT-O 10000 10100 10103 PR UIVT-O BB RVO:67985556 RVO:67985807 https://hdl.handle.net/11104/0356122 cav_un_auth*0472963 Technická Univerzita v Liberci, Ústav informatiky a výpočetní techniky AV TUL, UIVT CZ S 2024 C 978-3-031-65992-8 SCOPUS PUBMED 001323540900024 WOS cav_un_epca*0598452 Combining, Modelling and Analyzing Imprecision, Randomness and Dependence Springer 2024 Cham 200 207 978-3-031-65992-8 Advances in Intelligent Systems and Computing 2194-5357