Geodesic
Strange Design Points in Linear Regression
Acta Universitatis Palackianae Olomucensis. Facultas Rerum Naturalium. Mathematica, Tome 50 (2011) no. 2, pp. 103-110 Cet article a éte moissonné depuis la source Czech Digital Mathematics Library

Voir la notice de l'article

We discuss, partly on examples, several intuitively unexpected results in a standard linear regression model. We demonstrate that direct observations of the regression curve at a given point can not be substituted by observations at two very close neighboring points. On the opposite, we show that observations at two distant design points improve the variance of the estimator. In an experiment with correlated observations we show somewhat unexpected conditions under which a design point gives no or very little information about the estimated parameters, and so can be excluded from the design. For completeness we repeat briefly known conditions under which a design point is sensitive to the presence of outliers.
We discuss, partly on examples, several intuitively unexpected results in a standard linear regression model. We demonstrate that direct observations of the regression curve at a given point can not be substituted by observations at two very close neighboring points. On the opposite, we show that observations at two distant design points improve the variance of the estimator. In an experiment with correlated observations we show somewhat unexpected conditions under which a design point gives no or very little information about the estimated parameters, and so can be excluded from the design. For completeness we repeat briefly known conditions under which a design point is sensitive to the presence of outliers.
Classification : 62J05, 62K05
Keywords: singular models; optimal design; correlated observations 
@article{AUPO_2011_50_2_a10,
     author = {P\'azman, Andrej},
     title = {Strange {Design} {Points} in {Linear} {Regression}},
     journal = {Acta Universitatis Palackianae Olomucensis. Facultas Rerum Naturalium. Mathematica},
     pages = {103--110},
     year = {2011},
     volume = {50},
     number = {2},
     mrnumber = {2920712},
     zbl = {1244.62101},
     language = {en},
     url = {http://geodesic.mathdoc.fr/item/AUPO_2011_50_2_a10/}
}
TY  - JOUR
AU  - Pázman, Andrej
TI  - Strange Design Points in Linear Regression
JO  - Acta Universitatis Palackianae Olomucensis. Facultas Rerum Naturalium. Mathematica
PY  - 2011
SP  - 103
EP  - 110
VL  - 50
IS  - 2
UR  - http://geodesic.mathdoc.fr/item/AUPO_2011_50_2_a10/
LA  - en
ID  - AUPO_2011_50_2_a10
ER  - 
%0 Journal Article
%A Pázman, Andrej
%T Strange Design Points in Linear Regression
%J Acta Universitatis Palackianae Olomucensis. Facultas Rerum Naturalium. Mathematica
%D 2011
%P 103-110
%V 50
%N 2
%U http://geodesic.mathdoc.fr/item/AUPO_2011_50_2_a10/
%G en
%F AUPO_2011_50_2_a10
Pázman, Andrej. Strange Design Points in Linear Regression. Acta Universitatis Palackianae Olomucensis. Facultas Rerum Naturalium. Mathematica, Tome 50 (2011) no. 2, pp. 103-110. http://geodesic.mathdoc.fr/item/AUPO_2011_50_2_a10/

[1] Fišerová, E., Kubáček, L., Kunderová, P.: Linear Statistical Models: Regularity and Singularities. Academia, Praha, 2007.

[2] Harville, D. A.: Matrix Algebra from a Statistician’s Perspective. Springer, New York, 1997. | MR | Zbl

[3] Näther, W.: Exact designs for regression models with correlated errors. Statistics 16 (1985), 479–484. | DOI | MR

[4] Pázman, A.: Foundations of Optimum Experimentsl Design. Kluwer, Dordrecht, 1986.

[5] Pázman, A.: Information contained in design points of experiments with correlated observations. Kybernetika 46 (2010), 769–781. | MR | Zbl

[6] Pázman, A., Pronzato, L.: On the irregular behavior of LS estimators for asymptotically singular designs. Statistics and Probability Letters 76 (2006), 1089–1096. | DOI | MR | Zbl

[7] Pukelsheim, F.: Optimal Design of Experiments. Wiley, New York, 1993. | MR | Zbl

[8] Zvára, K.: Regresní analýza. Academia, Praha, 1989.