Abstract
Consider the design problem for the approximately linear model with serially correlated errors. The correlated structure is the qth degree moving average process, MA(q), especially for q = 1, 2. The optimal design is derived by using Bayesian approach. The Bayesian designs derived with various priors are compared with the classical designs with respect to some specific correlated structures. The results show that any prior knowledge about the sign of the MA(q) process parameters leads to designs that are considerately more efficient than the classical ones based on homoscedastic assumptions.
Similar content being viewed by others
References
Box GEP, Draper NR (1959) A basis for the selection of a response surface design. J Smer Stat Assoc 54: 622–654
Chang YJ, Notz WI (1996) Model Robust desings. In: Ghosh S, Rao CR(eds) Handbook of statistics, vol 13. Elsevier, Amsterdam, pp 1055–1098
Haines LM (1987) The application of the annealing algorithm to the construction of the exact optimaldesigns for linear-regression models. Technometrics 29: 439–447
Marcus MB, Sacks J (1978) Robust designs for regression problems. Statistical decision theory and related topics, II. Academic Press, New York, pp 45–268
Sacks J, Ylvisaker D (1978) Linear estimation for approximately linear models. Ann Stat 6: 1122–1137
Wiens DP (1992) Minimax design for approximately linear regression. J Stat Plann Inference 31: 353–371
Weins DP, Zhou JL (1996) Minimax desings for approximately linear models with autocorrelated errors. J Stat Plann Inference 92: 807–818
Weins DP, Zhou JL (1999) Minimax desings for approximately linear models with AR(1) errors. Can J Stat 27: 781–794
Yue RX (2002) Model-Robust design for response surface in \({\mathcal{R}^s}\) . Chin J Appl Probab Stat 18: 71–80
Yue RX, Hickernell FJ (1999) Robust designs for fitting linear models with misspecification. Stat Sin 9: 1053–1069
Yue RX, Wu JW (2004) U-type and factorial designs for nonparametric Bayesian regression. Stat Prob Lett 69: 343–356
Zhou J (2001a) Interger-valued, minimax robust designs fro approximately linear models with correlated errors. Commu Stat Theory Meth 30(1): 21–39
Zhou J (2001b) A robust criterion for experimental designs for serially correlated observations. Technometrics 43(4): 462–467
Zhou J (2001c) Two robust designs approaches for linear models cwith correlated errors. Stat Sin 11: 261–272
Author information
Authors and Affiliations
Corresponding author
Additional information
This work was partially supported by NSFC Grant (10671129), E-Institutes of Shanghai Municipal Education Commission (E03004), Special Funds for Doctoral Authorities of Education Ministry of China (20060270002), Science and Technology Commission of Shanghai Municipality grant 075105118, and Shanghai Leading Academic Discipline Project (5/Z/206).
Rights and permissions
About this article
Cite this article
Yue, RX., Zhou, XD. Bayesian robust designs for linear models with possible bias and correlated errors. Metrika 71, 1–15 (2010). https://doi.org/10.1007/s00184-008-0197-0
Received:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s00184-008-0197-0