• Marta Savkina Institute of Mathematics of NASU, Kyiv, Ukraine
Keywords: least square method, regression model, Aitken estimation


In the paper in the case of heteroscedastic independent deviations a regression model whose function has the form $ f (x) = ax^2 + bx + c$, where $a$, $b$, and $c$ are unknown parameters, is studied. Approximate values (observations) of functions $f (x)$ are registered at equidistant points of a line segment. The theorem proved in the paper states that Aitken estimation of the higher coefficient of the quadratic model in the case of odd the number of observation points coincides with its estimation of LS iff values of the variances satisfy a certain system of nonlinear equations. Under these conditions, the Aitken and LS estimations of $b$ and $c$ will not coincide. The application of the theorem for some cases of a specific quantity of observation points and the same values of the variances at nodes symmetric about the point $\frac{1}{2}$ is considered. In all these cases it is obtained that the LS estimation will be coincide Aitken estimation if the variance in two points accepts arbitrary values, and at all others does certain values that are expressed through the values of variances in these two points.


Anderson T. The statistical analysis of time series. Moscow: Mir, 1976. 756 p. (in Russian)

Demidenko E. Z. Linear and nonlinear regression. Moscow: Finance and Statistics, 1981. 304 p. (in Russian)

Grenander Ulf On the estimation of the regression coefficients in the case of the autocorrelated disturbance. Ann. Math. Statist. 1954. 25. P. 252–272.

Zyskind G. On cannonical forms, non-negative covariance matrices, and best and simple least squares linear estimators in linear models. Ann. Math. Statist. 1967. 38. P. 1092–1109.

Magness T. A., McGuire J. B. Comparison of least squares and minimum variance estimates of regression parameters. Ann. Math. Statist. 1962. 33. P. 462–470.

Rosenblatt M. Serial correlation in Some regression problems in the time series analysis. Proc. Third Berkelay Symposium on Mathematical Statistics and Probability (J. Neyman, ed.), Vol. 1, Univ. of California Press, Berkelay and Los Angeles. 1956. P. 165–186.

Watson G. S. Serial correlation in regression analysis, I. Biometrika. 1955. 42. P. 327–341.

Watson G. S. Linear least squares regression. Ann. Math. Statist. 1967. 38. P. 1679–1699.

Watson G. S., Hannan E. J. Serial correlation in regression analysis, II. Biometrika. 1956. 43. P. 436–449.

Belov A. G. About coincidence of the LS and Aitken estimations. Applied mathematics and Informatics. Proceedings of the faculty of the VMK Moscow State University. 2016. Vol. 51. P. 95–99. (in Russian)

Savkina M. Conditions for the coincidence of the IS and Aitken estimations of the parameters of the linear regression model. Journal of Numerical and Applied Mathematics. 2018. No. 3 (129). P. 36–44. (in Ukrainian)