Hamid Karamikabir

One of the most important issues in theoretical statistics is the problem of estimation which can be discussed in both Beysain and Classical approaches. Shrinkage estimation is one of the common methods that has attracted the attention of many researchers nowadays. In this thesis, the problem of shrinkage estimation of a multivariate mean vector is considered for multivariate normal distribution and more generally for scale mixtures of normal. The dominance of shrinkage estimators had been studied by comparing their risks under balance loss functions. Finally, all the theoretical results are studied using simulation study and real example in practice.