Journal of Statistics: Advances in Theory and ApplicationsAuthor's: Eiichi Isogai and Chikara Uno
Pages: [55] - [72]
Received Date: January 23, 2013
Submitted by:
In this paper, we consider the problem of minimum risk point estimation of the variance of a normal distribution when the loss function is squared error plus linear cost. We deal with the two-stage procedure proposed by Mukhopadhyay and Duggan [4] and provide a second-order asymptotic expansion of the risk. We also propose a bias-corrected procedure to reduce the risk and show that this procedure is effective in reducing the risk.
minimum risk, normal distribution, variance, two-stage procedure, second-order expansions, bias-correction.
