Abstract
Peaks over threshold model (POT) can describes the tail characteristics of unknown distribution. Threshold selection and parameter estimation is very important for POT model. This paper suggests kurtosis method for threshold selection. Genetic algorithm and pattern search is used to estimate the parameters of POT and then the estimated parameters by genetic algorithm are as the initial point of pattern search. VaR and CVaR are computed based on POT model. The experiment shows that POT model based on hybrid of genetic algorithm and pattern search is valid.
Full Text 
Sign-in/Register to access full text options
Sign-in/Register to access full text options 
Published version (
Free)
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
