Аннотация:In this paper, we consider the general linear stochastic model in which the variance-covariance matrix of the observations is a linear function of the variance-covariance components. Hence, first, we derive the formulas for the BIQUE (Best Invariant Quadratic Unbiased Estimates) estimation of the variance-covariance components in the functional model of the condition only. Next, from these particular formulas, the general procedure for the BIQUE variance-covariance components is derived, which can be applied to all least-squares adjustment functional models (condition only, Gauss-Markov, Gauss-Helmert, Gauss-Helmert with constraints among the parameters).
