This article is cited in 1 paper

Short Communications

On the choice of estimation technique for regression experimental designs unbiased in $L_2$-metric

E. V. Sedunov

Leningrad

Abstract: The paper deals with regression experimental designs unbiased in the $L_2$-metric. Consideration is concentrated upon the choice of estimation technique for design procedures. It is shown that the procedure using the reproducing kernel method dominates the least-squares procedure. In particular, results for polynominal and trigonometric regressions are derived. Numerical examples are given.

Received: 07.10.1974

 English version: Theory of Probability and its Applications, 1977, 21:2, 423–430

Bibliographic databases:

• 
•