The paper deals with an optimal estimation of the quadratic function $\bold{\beta'D\beta}$, where $\beta \in \Cal R^k, \bold D$ is a known $k \times k$ matrix, in the model $\bold{Y, X\beta, \sigma^2I}$. The distribution of $\bold Y$ is assumed to be symmetric and to have a finite fourth moment. An explicit form of the best unbiased estimator is given for a special case of the matrix $\bold X$.
@article{104343, author = {J\'ulia Volaufov\'a and Peter Volauf}, title = {Estimation of a quadratic function of the parameter of the mean in a linear model}, journal = {Applications of Mathematics}, volume = {34}, year = {1989}, pages = {155-160}, zbl = {0673.62053}, mrnumber = {0990302}, language = {en}, url = {http://dml.mathdoc.fr/item/104343} }
Volaufová, Júlia; Volauf, Peter. Estimation of a quadratic function of the parameter of the mean in a linear model. Applications of Mathematics, Tome 34 (1989) pp. 155-160. http://gdmltest.u-ga.fr/item/104343/
Simultaneous Estimation of Expectation and Covariance Matrix in Linear Models, Math. Operationsforsch. Statist., Ser. Statistics, Vol. 9 (1978) No. 3, 443-478. (1978) | MR 0522072 | Zbl 0415.62026
Estimation of Polynomials in the Regression Model, Aplikace matematiky, Vol. 27 (1982), No. 3, 223-231. (1982) | MR 0658004