A generalized class of estimators in linear regression models with multivariate-t distributed error

A generalized estimator representing a class of estimators is proposed for the estimation of regression coefficients in the linear regression model when the error components have the joint multivariate Student-t distribution. Approximate expressions for the bias and the risk of the proposed generalized estimator under a general quadratic loss function are found and a comparative study among some of the estimators of the class is made. A generalized efficiency (dominance) condition of the class over the usual minimum variance unbiased estimator (MVUE) is also given.