Abstract
The generalized Pareto model plays an important role in modelling extreme events. Hosking and Wallis (1987) discussed the parameter and quantile estimation for generalized Pareto distribution. Optimal experimental designs are used to accurately estimate the unknown parameters of the model. In this paper, locally D-, A- and E-optimal designs with two and three support points having equal and unequal weights for homoscedastic generalized Pareto regression model are obtained numerically. It is also proved that these designs are minimally supported. The results are illustrated through Norwegian fire insurance claim data.
Publisher
International Journal of Mathematical, Engineering and Management Sciences plus Mangey Ram
Subject
General Engineering,General Business, Management and Accounting,General Mathematics,General Computer Science
Reference19 articles.
1. Atkinson, A.C., Donev, A.N., & Tobias, R.D. (2007). Optimum experimental designs, with SAS. Oxford University Press. New York.
2. Beirlant, J., & Geogebeur, Y. (2003). Regression with response distributions of Pareto-type. Comutational Statistics & Data Analysis, 42(4), 595-619.
3. Castillo, J.D., & Daoudi, J. (2009). Estimation of generalized Pareto distribution. Statistics and Probability Letters, 79(5), 684-688.
4. Chernoff, H. (1953). Locally optimal designs for estimating parameters, The Annals of Mathematical Statistics, 24(4), 586–602.
5. Dette, H., & Haines, L.M. (1994). E-optimal designs for linear and nonlinear models with two parameters, Biometrika, 81(4), 739-754.