The paper discusses the problem of model risk, defined as risk resulting from the application of wrong model in real world. Three sources of model risk are distinguished: risk related to the structure of the model, risk of model estimation and risk connected with the application of the model. The main part of the paper presents the measures that can be used to evaluate risk of model estimation. Two particular cases are solved. The first one is the construction of two stock portfolio with minimal risk, the second one is option pricing. In both cases estimation risk results from the fact that main parameter, which is volatility (standard deviation if returns), has to be estimated. Finally, the paper states important conditions to limit model risk.
JavaScript jest wyłączony w Twojej przeglądarce internetowej. Włącz go, a następnie odśwież stronę, aby móc w pełni z niej korzystać.