BAYESIAN APPROXIMATIONS FOR SHAPEPARAMETER OF GENERALIZED POWER FUNCTION DISTRIBUTION

  • Hummara Sultan P.G Department of Statistics, University of Kashmir, Srinagar, India
  • S.P. Ahmad P.G Department of Statistics, University of Kashmir, Srinagar, India
Keywords: Bayesian Estimation, Prior Distribution, Normal Approximation, Lindley’s Approximation, Laplace Approximation

Abstract

Power function distribution provides a better fit for failure data and more appropriate information about reliability and hazard rates and used as a subjective description of a population for which there is limited sample data and in case where the relationship between variables is known but data is scare. In this paper, Bayesian approximation techniques like normal approximation, Lindley’s Approximation, Laplace Approximation are used to study the behavior of shape parameter of generalized power function distribution under different priors. Furthermore, a comparison of these approximation techniques, under different priors is studied by making use of simulation technique.

 

Downloads

Download data is not yet available.
Published
2017-06-12
Section
Articles