Volume no :9, Issue no: 2, June (2013)

Journal of Statistics: Advances in Theory and Applications

LOCAL PREDICTIVE ESTIMATES TOWARDS MLES FROM THE GENERALIZED HALF-NORMAL MODEL

Author's: HAFIZ M. R. KHAN, AHMED N. ALBATINEH and TARANG N. KHERADIA
Pages: [73] - [98]
Received Date: June 1, 2013
Submitted by:

Abstract

In this paper, the maximum likelihood function given a sample from the generalized half-normal model is derived. The maximum likelihood method is used to derive the predictive density for future responses. The predictive moments, skewness, and kurtosis for a single future response are determined. A comparison of the predictive variability of the maximum likelihood estimates (MLEs) and some of its local estimates are provided. Several scenarios of the predictive standard deviations for a single future response are determined and presented graphically. A real data set is used to illustrate the predictive results.

Keywords

statistical inference, generalized half-normal model, predictive inference.