Abstract
When the error distribution in a regression model is asymmetric, the second-order least squares estimator (SLSE) is more efficient than the ordinary least squares estimator. This result motivated the research in Gao and Zhou (J Stat Plan Inference 149:140–151, 2014), where A-optimal and D-optimal design criteria based on the SLSE were proposed and various design properties were studied. In this paper, we continue to investigate the optimal designs based on the SLSE and derive new results for the D-optimal designs. Using convex optimization techniques and moment theories, we can construct D-optimal designs for univariate polynomial and trigonometric regression models on any closed interval. Several theoretical results are obtained. The methodology is quite general. It can be applied to reduced polynomial models, reduced trigonometric models, and other regression models. It can also be extended to A-optimal designs based on the SLSE.
Similar content being viewed by others
References
Berger MPF, Wong WK (2009) An introduction to optimal designs for social and biomedical research. Wiley, Chichester
Boyd S, Vandenberghe L (2004) Convex optimization. Cambridge University Press, New York
Chang FC, Imhof L, Sun YY (2013) Exact D-optimal designs for first-order trigonometric regression models on a partial circle. Metrika 76:857–872
Curto RE, Fialkow LA (1991) Recursiveness, positivity, and truncated moment problems. Houst J Math 17:603–635
Dette H, Melas VB, Pepelyshev A (2002) D-optimal designs for trigonometric regression models on a partial circle. Ann Inst Stat Math 54:945–959
Dette H, Pepelyshev A, Zhigljavsky A (2008) Improving updating rules in multiplicative algorithms for computing D-optimal designs. Comput Stat Data Anal 53:312–320
Dette H, Studden WJ (1997) The theory of canonical moments with applications in statistics, probability, and analysis. Wiley, New York
Duarte BPM, Wong WK (2014a) A semi-infinite programming based algorithm for finding minimax D-optimal designs for nonlinear models. Stat Comput 24:1063–1080
Duarte BPM, Wong WK (2014b) Finding Bayesian optimal designs for nonlinear models: a semidefinite programming-based approach. Int Stat Rev (in press)
Duarte BPM, Wong WK, Atkinson AC (2015) A semi-infinite programming based algorithm for determining T-optimum designs for model discrimination. J Multivar Anal 135:11–24
Fedorov VV (1972) Theory of optimal experiments. Academic Press, New York
Gao LL, Zhou J (2014) New optimal design criteria for regression models with asymmetric errors. J Stat Plan Inference 149:140–151
Grant MC, Boyd SP (2013) The CVX Users’ Guide. Release 2.0 (beta), CVX Research, Inc. http://cvxr.com/cvx/doc/CVX. Accessed 14 Oct 2013
Huang MNL, Chang FC, Wong WK (1995) D-optimal designs for polynomial regression without an intercept. Stat Sin 5:441–458
Laurent M (2010) Updated version of Sums of squares, moment matrices and optimization over polynomials. http://homepages.cwi.nl/~monique/files/moment-ima-update-new. Accessed 6 Feb 2010
Lu Z, Pong TK (2013) Computing optimal experimental designs via interior point method. SIAM J Matrix Anal Appl 34:1556–1580
Papp D (2012) Optimal designs for rational function regression. J Am Stat Assoc 107:400–411
Pronzato L, Pázman A (2013) Design of experiments in nonlinear models—asymptotic normality, optimality criteria and small-sample properties. Springer, New York
Pukelsheim F (1993) Optimal design of experiments. Wiley, New York
Silvey SD, Titterington DM, Torsney B (1978) An algorithm for optimal designs on a finite design space. Commun Stat Theory Methods 14:1379–1389
Wang L, Leblanc A (2008) Second-order nonlinear least squares estimation. Ann Inst Stat Math 60:883–900
Xu X, Shang X (2014) Optimal and robust designs for trigonometric regression models. Metrika 77:753–769
Yu Y (2010) Monotonic convergence of a general algorithm for computing optimal designs. Ann Stat 38:1593–1606
Yu Y (2011) D-optimal designs via a cocktail algorithm. Stat Comput 21:475–481
Zhang C (2007) Optimal designs for trigonometric regression. Commun Stat Theory Methods 36:755–766
Acknowledgments
This research work is supported by Discovery Grants from the Natural Science and Engineering Research Council of Canada. The authors thank Professor Jiawang Nie for his valuable suggestions that lead to the development of Algorithm I. The authors are also grateful to the Editor and reviewers for their helpful comments and suggestions.
Author information
Authors and Affiliations
Corresponding author
Appendix: Proofs
Appendix: Proofs
Proof of Theorem 1
From (4), we have
\(\square \)
Proof of Theorem 3
-
(i)
For the \(\xi _D^{ SLS}\), it is clear that \(\phi (\mu _1^*, \ldots , \mu _{2q}^*) < \infty \), which implies that \(\det \left( \mathbf{A}(t, \mu _1^*, \ldots , \mu _{2q}^*) \right) > 0\). By (5), we have \(\det \left( \varvec{G}_2\right) > 0\), so the rank of \(\varvec{G}_2\) is \(q\). For model (7), since \( \varvec{G}_2 = \sum _{i=1}^N p_i \mathbf{f}(x_i^*) \mathbf{f}^\top (x_i^*)\), we must have \(N \ge q\), which means that there must be at least \(q\) support points. From Curto and Fialkow (1991) and Step (2 - i) in Algorithm I, there are at most \(q+1\) support points. Thus the number of support points is either \(q\) or \(q+1\).
-
(ii)
For a symmetric design space \(S=[-c, c]\), Gao and Zhou (2014) have proved that there exists a symmetric D-optimal design \(\xi _D^{ SLS}\). We prove by contradiction that the support points include the two boundary points. Suppose the ordered support points of a symmetric \(\xi _D^{ SLS}\) are \(x_1^* < x_2^* < \cdots < x_N^*\), and \(|x_1^*|=x_N^* < c\). Define \(\nu =\frac{c}{x_N^*}\), then we have \(\nu > 1\) and all the points \(\nu x_1^*, \nu x_2^*,\ldots , \nu x_N^*\) are still in the design space. Define a distribution \(\xi _D^+\) on \(S=[-c, c]\) as
$$\begin{aligned} \xi _D^{+} = \left( \begin{array}{c@{\quad }c@{\quad }c@{\quad }c} \nu x_1^* &{} \nu x_2^* &{} \cdots &{} \nu x_N^* \\ p_1 &{} p_2 &{} \cdots &{} p_N \end{array} \right) , \end{aligned}$$and its moments are \(\mu _j^+= \nu ^j \mu _j^*\), \(j=1, \ldots , 2q\). By (8), it is easy to verify that
$$\begin{aligned} \det \left( \mathbf{A}(t, \mu _1^+, \ldots , \mu _{2q}^+) \right)= & {} \nu ^{q(q+1)} \det \left( \mathbf{A}(t, \mu _1^*, \ldots , \mu _{2q}^*) \right) \\> & {} \det \left( \mathbf{A}(t, \mu _1^*, \ldots , \mu _{2q}^*) \right) , \end{aligned}$$which implies that \(\phi (\mu _1^+, \ldots , \mu _{2q}^+) < \phi (\mu _1^*, \ldots , \mu _{2q}^*)\). This is a contradiction to the fact that the \(\xi _D^{ SLS}\) minimizes \(\phi (\mu _1, \ldots , \mu _{2q})\). Therefore, the support points of \(\xi _D^{ SLS}\) must include the two boundary points \(-c\) and \(c\). The results in (iii) and (iv) can be proved similarly to the result in (ii), and the proof is omitted.
\(\square \)
Rights and permissions
About this article
Cite this article
Gao, L.L., Zhou, J. D-optimal designs based on the second-order least squares estimator. Stat Papers 58, 77–94 (2017). https://doi.org/10.1007/s00362-015-0688-9
Received:
Revised:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s00362-015-0688-9
Keywords
- Asymmetric distribution
- Convex optimization
- Moment theory
- Optimal design
- Polynomial regression
- Trigonometric regression