##
**\(T\)-optimal designs for discrimination between two polynomial models.**
*(English)*
Zbl 1246.62176

Summary: This paper is devoted to the explicit construction of optimal designs for discrimination between two polynomial regression models of degree \(n-2\) and \(n\). In a fundamental paper, A.C. Atkinson and V.V. Fedorov [Biometrika, 62, 57–70 (1975; Zbl 0308.62071)] proposed the \(T\)-optimality criterion for this purpose. A.C. Atkinson [The non-uniqueness of some designs for discriminating between two polynomial models in one variablel. mODa 9, Advances in Model-Oriented Design and Analysis, 9–16 (2010)] determined \(T\)-optimal designs for polynomials up to degree 6 numerically and based on these results he conjectured that the support points of the optimal design are cosines of the angles that divide half of the circle into equal parts if the coefficient of \(x^{n-1}\) in the polynomial of larger degree vanishes. We give a strong justification of the conjecture and determine all \(T\)-optimal designs explicitly for any degree \(n \in \mathbb N\).

In particular, we show that there exists a one-dimensional class of \(T\)-optimal designs. Moreover, we also present a generalization to the case when the ratio between the coefficients of \(x^{n-1}\) and \(x^{n}\) is smaller than a certain critical value. Because of the complexity of the optimization problem, \(T\)-optimal designs have only been determined numerically so far, and this paper provides the first explicit solution of the \(T\)-optimal design problem since its introduction by Atkinson and Fedorov. Finally, for the remaining cases (where the ratio of coefficients is larger than the critical value), we propose a numerical procedure to calculate the \(T\)-optimal designs. The results are also illustrated by an example.

In particular, we show that there exists a one-dimensional class of \(T\)-optimal designs. Moreover, we also present a generalization to the case when the ratio between the coefficients of \(x^{n-1}\) and \(x^{n}\) is smaller than a certain critical value. Because of the complexity of the optimization problem, \(T\)-optimal designs have only been determined numerically so far, and this paper provides the first explicit solution of the \(T\)-optimal design problem since its introduction by Atkinson and Fedorov. Finally, for the remaining cases (where the ratio of coefficients is larger than the critical value), we propose a numerical procedure to calculate the \(T\)-optimal designs. The results are also illustrated by an example.

### MSC:

62K05 | Optimal statistical designs |

65C60 | Computational problems in statistics (MSC2010) |

### Citations:

Zbl 0308.62071### References:

[1] | Achiezer, N. I. (1956). Theory of Approximation . Ungar, New York. · Zbl 0074.45002 |

[2] | Asprey, S. P. and Macchietto, S. (2000). Statistical tools for optimal dynamic model building. Computers and Chemical Engineering 24 1261-1267. |

[3] | Atkinson, A. C. (2010). The non-uniqueness of some designs for discriminating between two polynomial models in one variablel. In mODa 9, Advances in Model-Oriented Design and Analysis (A. Giovagnoli, A. C. Atkinson, B. Torsney and C. May, eds.) 9-16. Physica, Heidelberg. |

[4] | Atkinson, A. C., Bogacka, B. and Bogacki, M. B. (1998). D - and T -optimum designs for the kinetics of a reversible chemical reaction. Chemometrics and Intelligent Laboratory Systems 43 185-198. |

[5] | Atkinson, A. C. and Fedorov, V. V. (1975a). The design of experiments for discriminating between two rival models. Biometrika 62 57-70. · Zbl 0308.62071 |

[6] | Atkinson, A. C. and Fedorov, V. V. (1975b). Optimal design: Experiments for discriminating between several models. Biometrika 62 289-303. · Zbl 0321.62085 |

[7] | Chaloner, K. and Verdinelli, I. (1995). Bayesian experimental design: A review. Statist. Sci. 10 273-304. · Zbl 0955.62617 |

[8] | Chernoff, H. (1953). Locally optimal designs for estimating parameters. Ann. Math. Statist. 24 586-602. · Zbl 0053.10504 |

[9] | Denisov, V. I., Fedorov, V. V. and Khabarov, V. I. (1981). Tchebysheff approximation in problems of constructing asymptotic locally optimal designs for discriminating regression experiments. In Problems in Cibernetics ( Voprosy Kibernetiki in Russian ), Linear and Non-Linear Parametrization in Optimal Design (V. Fedorov and V. Nalimov, eds.) 3. Akademiya Nauk RSFSR, Nauchnyj Sovet po Kompleksnoj Probleme Kibernetika, Moskva. · Zbl 0489.62066 |

[10] | Dette, H. (1994). Discrimination designs for polynomial regression on compact intervals. Ann. Statist. 22 890-903. · Zbl 0806.62059 |

[11] | Dette, H. (1995). Optimal designs for identifying the degree of a polynomial regression. Ann. Statist. 23 1248-1266. · Zbl 0847.62064 |

[12] | Dette, H. (1997). Designing experiments with respect to “standardized” optimality criteria. J. Roy. Statist. Soc. Ser. B 59 97-110. · Zbl 0884.62081 |

[13] | Dette, H. and Haller, G. (1998). Optimal designs for the identification of the order of a Fourier regression. Ann. Statist. 26 1496-1521. · Zbl 0930.62075 |

[14] | Dette, H., Melas, V. B. and Pepelyshev, A. (2004). Optimal designs for estimating individual coefficients in polynomial regression-a functional approach. J. Statist. Plann. Inference 118 201-219. · Zbl 1031.62055 |

[15] | Dette, H. and Titoff, S. (2009). Optimal discrimination designs. Ann. Statist. 37 2056-2082. · Zbl 1168.62066 |

[16] | Fedorov, V. (1980). Design of model testing experiments. In Symposia Mathematica , Vol. XXV ( Conf. , INDAM , Rome , 1979) 171-180. Academic Press, London. · Zbl 0468.62071 |

[17] | Fedorov, V. and Khabarov, V. (1986). Duality of optimal designs for model discrimination and parameter estimation. Biometrika 73 183-190. · Zbl 0597.62082 |

[18] | Foo, L. K. and Duffull, S. (2011). Optimal design of pharmacokinetic-pharmacodynamic studies. In Pharmacokinetics in Drug Development , Advances and Applications 3 (P. L. Bonate and D. R. Howard, eds.) 175-194. Springer, Berlin. |

[19] | Hill, P. D. (1978). A review of experimental design procedures for regression model discrimination. Technometrics 20 15-21. · Zbl 0378.62067 |

[20] | Hunter, W. G. and Reiner, A. M. (1965). Designs for discriminating between two rival models. Technometrics 7 307-323. |

[21] | Jolley, L. B. W. (1961). Summation of Series , 2nd revised ed. Dover, New York. · Zbl 0101.28602 |

[22] | Kiefer, J. (1974). General equivalence theory for optimum designs (approximate theory). Ann. Statist. 2 849-879. · Zbl 0291.62093 |

[23] | Melas, V. B. (2006). Functional Approach to Optimal Experimental Design. Lecture Notes in Statistics 184 . Springer, New York. · Zbl 1094.62086 |

[24] | Song, D. and Wong, W. K. (1999). On the construction of G rm -optimal designs. Statist. Sinica 9 263-272. · Zbl 0913.62069 |

[25] | Spruill, M. C. (1990). Good designs for testing the degree of a polynomial mean. Sankhyā Ser. B 52 67-74. · Zbl 0733.62075 |

[26] | Stigler, S. (1971). Optimal experimental design for polynomial regression. J. Amer. Statist. Assoc. 66 311-318. · Zbl 0217.51701 |

[27] | Studden, W. J. (1980). D s -optimal designs for polynomial regression using continued fractions. Ann. Statist. 8 1132-1141. · Zbl 0447.62070 |

[28] | Studden, W. J. (1982). Some robust-type D -optimal designs in polynomial regression. J. Amer. Statist. Assoc. 77 916-921. · Zbl 0505.62062 |

[29] | Szegő, G. (1975). Orthogonal Polynomials , 4th ed. American Mathematical Society , Colloquium Publications XXIII . Amer. Math. Soc., Providence, RI. · Zbl 0305.42011 |

[30] | Tommasi, C. and López-Fidalgo, J. (2010). Bayesian optimum designs for discriminating between models with any distribution. Comput. Statist. Data Anal. 54 143-150. · Zbl 1284.62481 |

[31] | Uciński, D. and Bogacka, B. (2005). T -optimum designs for discrimination between two multiresponse dynamic models. J. R. Stat. Soc. Ser. B Stat. Methodol. 67 3-18. · Zbl 1060.62084 |

[32] | Wiens, D. P. (2009). Robust discrimination designs. J. R. Stat. Soc. Ser. B Stat. Methodol. 71 805-829. · Zbl 1248.62132 |

[33] | Wiens, D. P. (2010). Robustness of design for the testing of lack of fit and for estimation in binary response models. Comput. Statist. Data Anal. 54 3371-3378. · Zbl 1284.62492 |

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. It attempts to reflect the references listed in the original paper as accurately as possible without claiming the completeness or perfect precision of the matching.