CLOSED-FROM BAYESIAN ESTIMATION AND PREDICTION IN A NEW BIVARIATE RAYLEIGH MODEL USING JEFFREYS’ PRIOR

This paper addresses the problem of obtaining closed-form Bayesian point estimations and predictive density estimating from a new bivariate Rayleigh model using Jeffreys’ prior. In the presence of additional information, we propose some explicit estimators that outperform under the squared error and the normalized squared loss functions. We also find two closed-form density estimators that dominate the plug-in density estimators under the Kullback Leibler loss function. We show that the obtained predictive density estimator using additional information dominates the one ignoring this information. Finally, a real example applied to the hydrological flood data is considered to illustrate all the methods of inference discussed here.