D-optimal designs of experiments for trigonometric regression on interval with heteroscedastic observations

  • Valery P. Kirlitsa Belarusian State University, 4 Niezaliežnasci Avenue, Minsk 220030, Belarus

Abstract

In article the problem of construction continuous (number of observations is not fixed) D-optimal designs of experiments for trigonometric regression in a case when variance of errors of observations depend on a point in which is made is investigated. Class of functions which describe change variance of heteroscedastic observations is defined for which it is possible construct continuous D-optimal designs of experiments. For trigonometric regression with three factors it is constructed continuous D-optimal designs of experiments with different types heteroscedastic observations. For each of these types the own class of functions describing change variance of observations is defined.

Author Biography

Valery P. Kirlitsa, Belarusian State University, 4 Niezaliežnasci Avenue, Minsk 220030, Belarus

PhD (physics and mathematics), docent; associate professor at the department of mathematical modeling and data analysis, faculty of applied mathematics and computer science

References

  1. Ermakov SM, Zhiglyavskii AA. Matematicheskaya teoriya optimal’nogo eksperimenta [The mathematical theory of optimal design]. Moscow: Nauka; 1987. 320 p. Russian.
  2. Fedorov VV. Teoriya optimal’nogo eksperimenta [Optimal experiment theory]. Moscow: Nauka; 1971. 312 p. Russian.
  3. Hoel PG. Minimax designs in two dimensional regression. Annals of Mathematical Statistics. 1965;36(4):1097–1106. DOI: 10.1214/aoms/1177699984.
  4. Dette H, Melas VB. Optimal designs for estimating individual coefficients in Fourier regression models. Annals of Statistics. 2003;31(5):1669–1692. DOI: 10.1214/aos/1065705122.
Published
2020-12-08
Keywords: continuous D-optimal designs of experiments, trigonometric regression, homoscedastic observations, heteroscedastic observations
How to Cite
Kirlitsa, V. P. (2020). D-optimal designs of experiments for trigonometric regression on interval with heteroscedastic observations. Journal of the Belarusian State University. Mathematics and Informatics, 3, 80-85. https://doi.org/10.33581/2520-6508-2020-3-80-85