An Efficient Stochastic Approximation Algorithm and Model Selection for a Class of Nonlinear Random Effect Models with Incomplete Responses

Lin Dai; Ying Zi Fu

doi:10.4028/www.scientific.net/AMM.110-116.2670

Paper Titles

Numerical Modeling and Design of Thermoelectric Cooling Systems
p.2639

Modeling and Control of Ultra Precision Positioning System for a Grating Ruling Machine
p.2647

A Bayesian Method for Model Optimization in Structural Equation Models with Mixed Continuous and Ordered Categorical Data
p.2655

Optimization of Highly Non-Linear Simulation Functions Using Improved Adaptive Response Surface Methodology
p.2662

An Efficient Stochastic Approximation Algorithm and Model Selection for a Class of Nonlinear Random Effect Models with Incomplete Responses
p.2670

A New Approach for Simulation of Annular Two-Phase Flow Considering the Four Involved Mass Transfer Mechanisms (Annalytical Concept)
p.2677

FEA of Residual Stresses in Butt Welded Type Low Carbon Steel Using MMAW Technique
p.2686

Sedimentation Load Analysis Using ANN and GA
p.2693

An Interpretive Structural Modeling Approach for Analysis of Interactions among the Barriers in Reverse Logistics
p.2699

HomeApplied Mechanics and MaterialsApplied Mechanics and Materials Vols. 110-116An Efficient Stochastic Approximation Algorithm...

An Efficient Stochastic Approximation Algorithm and Model Selection for a Class of Nonlinear Random Effect Models with Incomplete Responses

Abstract:

A stochastic approximation expectation maximum (EM) algorithm is proposed to obtain Maximum likelihood (ML) estimation of nonlinear random effect models in which the manifest variables are distributed as a reproductive dispersion model (RDM) and may be missing with ignorable missingness mechanism in this paper. A method composed of simulation step as well as stochastic approximation step is used to obtain the conditional expectation, whereas the M-step is executed via the method of conditional maximization. The most attractive point of this approach is that it is novel and non-trivial, which can be used to obtain the ML estimates and the estimation of standard errors simultaneously. Moreover, A model selection criterion is developed. Empirical results are used to illustrate the usefulness of the methodologies proposed above.

You might also be interested in these eBooks

Mechanical and Aerospace Engineering, ICMAE2011

View Preview

Info:

Periodical:

Applied Mechanics and Materials (Volumes 110-116)

Pages:

2670-2676

DOI:

https://doi.org/10.4028/www.scientific.net/AMM.110-116.2670

Citation:

Cite this paper

Online since:

October 2011

Authors:

Lin Dai, Ying Zi Fu

Keywords:

Ignorable Missing Data, Model Selection, Nonlinear Random Effect Models, Random Effect

Export:

RIS, BibTeX

Price:

Permissions CCC:

Request Permissions

Permissions PLS:

Request Permissions

Сopyright:

Citation:

References

[1] S. Walker, An EM algorithm for nonlinear random effects model, Biometrics, Vol. 52, pp.934-944, (1993).

Google Scholar

[2] T. L. Lai and M. C. Shih, A hybrid estimator in nonlinear and generalized linear mixed effects models, Biometrics, Vol. 38, pp.859-879, (2003).

Google Scholar

[3] R. J. A. Little and D. B. Rubin, Statistical analysis with missing data (2nd ed. ). Wiley, (2002).

Google Scholar

[4] J. G. Ibrahim and M. H. Chen, and S. R. Lipsitz, Monte Carlo EM for missing covriates in parametric regression models, Biometrics, Vol. 55, pp.591-596, (1999).

DOI: 10.1111/j.0006-341x.1999.00591.x

Google Scholar

[5] J. G. Ibrahim and M. H. Chen, and S. R. Lipsitz, Missing responses in generalized linear mixed models when the missing data mechanism is nonignorable, Biometrika, Vol. 88, pp.551-564, (2001).

DOI: 10.1093/biomet/88.2.551

Google Scholar

[6] B. Jorgensen, The theory of dispersion models. London, Chapman and Hall, (1997).

Google Scholar

[7] G. O. Roberts, Markov chain concepts related to sampling algorithm, In: W. R. Gilks, S. Richardson, D. J. Spiegelhalter (ED. ), Markov chain Monte Carlo in Practice, London, Chapman and Hall, (1996).

DOI: 10.1201/b14835

Google Scholar

[8] X. L. Meng and W. H. Wong, Simulating ratios of normalizing constants via a simple identity: A theoretical exploration, Statistica Sinica, Vol. 6, pp.831-860, (1996).

Google Scholar

[9] T. A. Louis, Finding the observed information matrix when using EM algorithm, Journal of the Royal Statistical Association Series B, Vol. 44, pp.226-233, (1982).

DOI: 10.1111/j.2517-6161.1982.tb01203.x

Google Scholar

[10] J. G. Ibrahim, H.T. Zhu, and N. S. Tang, An AIC-type model selection criterion for missing data problem, Journal of American Statistical Association, Vol. 103, pp.1648-1658, (2008).

DOI: 10.1198/016214508000001057

Google Scholar