×

On a mixture of the \(D\)- and \(D_ 1\)-optimality criterion in polynomial regression. (English) Zbl 0768.62056

Summary: We consider a mixture of the \(D_ 1\)-optimality criterion (minimizing the variance of the estimate for the highest coefficient) and the \(D\)- optimality criterion (minimizing the volume of the ellipsoid of concentration for the unknown parameter vector) in the polynomial regression model of degree \(n\in \mathbb{N}\). The mixture is defined as a weighted product of both optimality criteria and explicit solutions are given for the proposed criterion. The derived designs have excellent efficiencies compared to the \(G\)-, \(D\)- and \(D_ 1\)-optimal design.
This is illustrated in some examples for polynomial regression of lower degree. The optimal designs are calculated using the theory of canonical moments. Further applications are given in the field of model robust designs for polynomial regression models where only an upper bound for the degree of the polynomial regression model is known by the experimenter before the experiments are carried out.

MSC:

62K05 Optimal statistical designs
62J02 General nonlinear regression
62J05 Linear regression; mixed models
Full Text: DOI

References:

[1] Abramowitz, M.; Stegun, I. A., Handbook of mathematical functions (1964), Dover: Dover New York · Zbl 0515.33001
[2] Atkinson, A. C., Planning experiments to detect inadequate regression models, Biometrika, 59, 275-293 (1972) · Zbl 0243.62046
[3] Cook, R. D.; Nachtsheim, C. J., Model robust, linear optimal designs, Technometrics, 24, 49-54 (1982) · Zbl 0483.62063
[4] Dette, H., Optimale Versuchspläne für mehrere konkurriende Polynommodelle bei einer gegebenen a-priori Gewichtung, Dissertation Universität Hannover (1989), (in German) · Zbl 0681.62060
[5] Dette, H., A generalization of \(D\)- and \(D_1\)-optimal design in polynomial regression, Ann. Statist., 18, 1784-1804 (1990) · Zbl 0714.62068
[6] Dette, H., A note on model robust design in polynomial regression, J. Statist. Plann. Inference, 28, 223-232 (1991) · Zbl 0728.62072
[7] Dette, H., Optimal designs for a class of polynomials of odd or even degree, Ann. Statist., 20, 238-259 (1992) · Zbl 0745.62075
[8] Fedorov, V. V., Theory of Optimal Experiments (1972), Academic Press: Academic Press New York
[9] Gutmair, S., Polar and subgradients of mixtures of information functions, Technical Report No. 286 (1991), Universität Augsburg, Schwerpunktprogramm der DFG ‘Anwendungsbezogene Optimierung und Steuerung’
[10] Hoel, P. G., Efficiency problems in polynomial estimation, Ann. Math. Statist., 29, 1134-1145 (1958) · Zbl 0094.14501
[11] Kiefer, J.; Wolfowitz, J., Optimum designs in regression problems, Ann. Math. Statist., 30, 271-294 (1959) · Zbl 0090.11404
[12] Kiefer, J.; Wolfowitz, J., The equivalence of two extremum problems, Canad. J. Math., 12, 363-366 (1960) · Zbl 0093.15602
[13] Läuter, E., Experimental design in a class of models, Math. Oper. Statist., 5, 379-398 (1974) · Zbl 0297.62056
[14] Läuter, E., Optimal multipurpose designs for regression models, Math. Oper. Statist., 7, 51-68 (1976) · Zbl 0329.62060
[15] Lau, T. S., Theory of canonical moments and its application in polynomial regression, I and II, Technical Reports 83-23 and 83-24 (1983), Dept. of Statistics, Purdue University
[16] Lau, T. S., \(D\)-optimal designs on the unit \(q\)-ball, J. Statist. Plann. Inference, 19, 299-315 (1988) · Zbl 0850.62603
[17] Lau, T. S.; Studden, W. J., Optimal designs for trigonometric and polynomial regression using canonical moments, Ann. Statist., 13, 383-394 (1985) · Zbl 0592.62067
[18] Lau, T. S.; Studden, W. J., On an extremal problem of Fejér, J. Approx. Theory, 53, 184-194 (1988) · Zbl 0663.41007
[19] Lim, Y. B.; Studden, W. J., Efficient \(D_s\)-optimal design for multivariate polynomial regression on the \(q\)-cube, Ann. Statist., 16, 1225-1240 (1988) · Zbl 0664.62075
[20] Perron, O., Die Lehre von den Kettenbrüchen (1954), Teubner: Teubner Stuttgart, (Band I&II) · JFM 55.0262.09
[21] Silvey, S. D., Optimal Design (1980), Chapman and Hall: Chapman and Hall London · Zbl 0468.62070
[22] Skibinsky, M., The range of the \((n + 1)\)-th moment for distributions on [0, 1], J. Appl. Probab., 4, 543-552 (1967) · Zbl 0189.18803
[23] Skibinsky, M., Extreme \(n\)-th moments for distributions on [0, 1] and the inverse of a moment space map, J. Appl. Probab., 5, 693-701 (1968) · Zbl 0176.48203
[24] Skibinsky, M., Some striking properties of binomial and beta moments, Ann. Math. Statist., 40, 1753-1764 (1969) · Zbl 0182.52401
[25] Skibinsky, M., Principal representations and canonical moment sequences for distribution on an interval, J. Math. Anal. Appl., 120, 95-120 (1986) · Zbl 0607.60013
[26] Stigler, S. M., Optimal experimental design for polynomial regression, J. Amer. Statist. Assoc., 66, 311-318 (1971) · Zbl 0217.51701
[27] Studden, W. J., \(D_s\)-optimal designs for polynomial regression using continued fractions, Ann. Statist., 8, 1132-1141 (1980) · Zbl 0447.62070
[28] Studden, W. J., Some robust-type \(D\)-optimal designs in polynomial regressions, J. Amer. Statist. Assoc., 77, 916-921 (1982) · Zbl 0505.62062
[29] Studden, W. J., Optimal designs for weighted polynomial regression using canonical moments, (Statistical Decision Theory and Related Topics III (1982), Academic Press: Academic Press New York), 335-350 · Zbl 0574.62067
[30] Studden, W. J., None on some \(Φ_p\)-optimal design for polynomial regression, Ann. Statist., 17, 618-623 (1989) · Zbl 0675.62047
[31] Szegö, G., Orthogonal Polynomials. American Mathematical Society Colloquium Publications, No. 23 (1959) · Zbl 0089.27501
[32] Wall, H. S., Analytic Theory of Continued Fractions (1948), Van Nostrand: Van Nostrand New York · Zbl 0035.03601
This reference list is based on information provided by the publisher or from digital mathematics libraries. Its items are heuristically matched to zbMATH identifiers and may contain data conversion errors. In some cases that data have been complemented/enhanced by data from zbMATH Open. This attempts to reflect the references listed in the original paper as accurately as possible without claiming completeness or a perfect matching.