Bayes estimate of the probability of exceedance of annual floods |
| |
Authors: | L M Lye |
| |
Institution: | (1) Faculty of Engineering and Applied Science, Memorial University of Newfoundland, A1B 3X5 St. John's, Newfoundland, Canada |
| |
Abstract: | In this paper Lindley's Bayesian approximation procedure is used to obtain the Bayes estimate of the probability of exceedence of a flood discharge. The Bayes estimates of the probability of exceedence has been shown by S.K. Sinha to be equivalent to the estimate of the probability of exceedence from the predictive or Bayesian disribution, of a future flood discharge. The evaluation of complex ratios of multiple integrals common in a Bayesian analysis is not necessary using Lindley's procedure. The Bayes estimates are compared to those obtained by the method of maximum likelihood and the method of moments. The results show that Bayes estimates of the probability of exceedence are larger as expected, but have smaller posterior standard deviations. |
| |
Keywords: | Predictive distribution Bayesian approximation parameter uncertainty non-informative prior method of moments Gumbel distribution maximum likelihood estimates |
本文献已被 SpringerLink 等数据库收录! |