AbstractThis article surveys the most important developments in volatility forecast comparison and model selection. We review a number of evaluation methods and testing procedures for predictive accuracy based on statistical loss functions. We also review recent contributions on the admissible form of loss functions ensuring consistency of the ordering when forecast performances are evaluated with respect to an imperfect volatility proxy. The techniques discussed are illustrated using artificial and EUR/USD exchange rate data. WIREs Comp Stat 2012, 4:1–12. doi: 10.1002/wics.190This article is categorized under: Statistical Models > Time Series Models Data: Types and Structure > Time Series, Stochastic Processes, and Functional Data Applications of Computational Statistics > Computational Finance