We suppose that a constraint is imposed on the value of the estimator in Bayesian Estimation Theory and study the distribution of this estimator under various hypotheses, using the two main loss functions: the absolute value and the quadratic.