E-Optimal Designs for Polynomial Regression
Pukelsheim, Friedrich ; Studden, William J.
Ann. Statist., Tome 21 (1993) no. 1, p. 402-415 / Harvested from Project Euclid
Text on Project Euclid
PDF
Alt PDF
E-optmal designs for the full mean parameter vector, and for many subsets in univariate polynomial regression models are determined. The derivation is based on the interplay between E-optimality and scalar optimality. The scalar parameter systems are obtained as transformations of the coefficient vector c of the Chebyshev polynomial.
Publié le : 1993-03-14
Classification:  Approximate design theory,  Chebyshev polynomial,  c-optimality,  E-optimality,  parameter subset optimality,  polynomial regression,  62K05 
@article{1176349033,
     author = {Pukelsheim, Friedrich and Studden, William J.},
     title = {E-Optimal Designs for Polynomial Regression},
     journal = {Ann. Statist.},
     volume = {21},
     number = {1},
     year = {1993},
     pages = { 402-415},
     language = {en},
     url = {http://dml.mathdoc.fr/item/1176349033}
}

Pukelsheim, Friedrich; Studden, William J. E-Optimal Designs for Polynomial Regression. Ann. Statist., Tome 21 (1993) no. 1, pp.  402-415. http://gdmltest.u-ga.fr/item/1176349033/