Empirical risk minimization for heavy-tailed losses
نویسندگان
چکیده
The purpose of this paper is to discuss empirical risk minimization when the losses are not necessarily bounded and may have a distribution with heavy tails. In such situations usual empirical averages may fail to provide reliable estimates and empirical risk minimization may provide large excess risk. However, some robust mean estimators proposed in the literature may be used to replace empirical means. In this paper we investigate empirical risk minimization based on a robust estimate proposed by Catoni. We develop performance bounds based on chaining arguments tailored to Catoni’s mean estimator.
منابع مشابه
Reduced-bias estimators for the Distortion Risk Premiums for Heavy-tailed distributions
Estimation of the occurrence of extreme events actually is that of risk premiums interest in actuarial Sciences, Insurance and Finance. Heavy-tailed distributions are used to model large claims and losses. In this paper we deal with the empirical estimation of the distortion risk premiums for heavy tailed losses by using the extreme value statistics. This approach can produce a potential bias i...
متن کاملOn optimal portfolio diversification with respect to extreme risks
Extreme losses of portfolios with heavy-tailed components are studied in the framework of multivariate regular variation. Asymptotic distributions of extreme portfolio losses are characterized by a functional γξ = γξ(Ψ, α) of the tail index α, the spectral measure Ψ, and the vector ξ of portfolio weights. Existence, uniqueness, and location of the optimal portfolio are analysed and applied to t...
متن کاملAggregation Issues in Operational Risk
In this paper we study copula-based models for aggregation of operational risk capital across business lines in a bank. A commonly used method of summation of the value-at-risk (VaR) measures, that relies on a hypothesis of full correlation of losses, becomes inappropriate in the presence of dependence between business lines and may lead to over-estimation of the capital charge. The problem can...
متن کاملOptimal oracle inequalities for model selection
Abstract: Model selection is often performed by empirical risk minimization. The quality of selection in a given situation can be assessed by risk bounds, which require assumptions both on the margin and the tails of the losses used. Starting with examples from the 3 basic estimation problems, regression, classification and density estimation, we formulate risk bounds for empirical risk minimiz...
متن کاملLearning sub-Gaussian classes : Upper and minimax bounds
Most the results contained in this note have been presented at the SMF meeting, which took place in May 2011; the rest have been obtained shortly after the time of the meeting. The question we study has to do with the optimality of Empirical Risk Minimization as a learning procedure in a convex class – when the problem is subgaussian. Subgaussian learning problems are a natural object because t...
متن کامل