David van Dyk's Home Page

David van Dyk, Associate Professor
Department of Statistics
University of California, Irvine

US Mail address:
346B ICS Bldg 1
Irvine, CA 92697-1250
Phone: 949-824-5679
Fax: 949-824-1683

e-mail: d v d [at] u c i [dot] e d u

Last modified July 30, 2003

David van Dyk is currently an associate professor in the Department of Statistics at The University of California at Irvine. Professor van Dyk received his Ph.D. from the Department of Statistics at the University of Chicago and his B.S. in Statistics and Probability from Michigan State University. His scholarly work focuses on methodological and computational issues involved with Bayesian analysis of highly structured statistical models and emphasizes serious interdisciplinary research. He is particularly interested in improving the efficiency of computationally intensive methods involving data augmentation, such as EM-type algorithms and various Markov chain Monte Carlo methods. Professor van Dyk's primary area of interdisciplinary work is Astro-statistics; he coordinates the California-Harvard Astrostatisics Collaboration, The group includes Astronomers and Statisticians and provides an open forum for discussing statistical issues that arise in astrophysical research, for example at the Harvard-Smithsonian Astrophysical Observatory. The group's current work focuses on constructing and fitting highly structured models for data obtained with the Chandra X-ray Observatory.

Research Links:

The homepage for the California-Harvard Astrostatisics Collaboration
A review paper summarizing recent advances in EM-type and Data Augmentation algorithms. A good place read about some of the basics of Professor van Dyk's research. The paper is available as a PostScript File or a PDF File.

Teaching Links:

Statistics 7: Basic Statistics, Syllabus.
Statistics 120A: Introduction to Probability and Statistics, Syllabus.
Statistics 200A: Intermediate Probability Theory, Syllabus.
Statistics 230: Statistical Computing Methods, Syllabus.

Publications and Manuscripts

In Refereed Journals (and current submissions)

van Dyk, D. A., Connors, A., Esch, D. Freeman, P., Kang, H., Karovska, M., Kashyap, V., Siemiginowska, A., and Zezas, A. (2004). Bayesian Deconvolusion Methods for Spectral and Image Analysis in High Energy Astrophysics with Discussion. Submitted to Bayesian Analysis .
Imai, K. and van Dyk, D. A. (2003). Causal Inference with General Treatments Regimes: Generalizing the Propensity Score. Revised for the Journal of the American Statistical Association.
van Dyk, D. A. and Kang, H. (2003). Highly Structured Hierarchical Models for Spectral Analysis in High Energy Astrophysics. Invited paper for special astro-statistics issue of Statistical Science , to appear.
Imai, K. and van Dyk, D. A. (2003). A Bayesian Analysis of the Multinomial Probit Model Using Marginal Augmentation. Submitted to Journal of Econometrics. abstract.
Javaras, K. N. and van Dyk, D. A. (2003). Multiple Imputation for Incomplete Data with Semicontinuous Variables. Journal of the American Statistical Association, to appear. abstract.
van Dyk, D. A. and Tang, R. (2003). The One-Step-Late PXEM Algorithm. Statistics and Computing, 13, 137-152. abstract.
Protassov, R., van Dyk, D. A., Connors, A., Kashyap, V. L. and Siemiginowska, A. (2002). Statistics: Handle with Care, Detecting Multiple Model Components with the Likelihood Ratio Test. The Astrophysical Journal, 571, 545-559.
van Dyk, D. A. (2002). Fast new EM-type algorithms with applications in Astrophysics. In revision for The Astrophysical Journal, abstract.
van Dyk, D. A., Connors, A., Kashyap, V. L., Siemiginowska, A. (2001). Analysis of Energy Spectra with Low Photon Counts via Bayesian Posterior Simulation. The Astrophysical Journal , 548, 224-243.
van Dyk, D. A. and Meng, X. L. (2001). The art of data augmentation (with discussion). Journal of Computational and Graphical Statistics, 10, 1-111, abstract.
Foully, J. L., van Dyk, D. A. (2000). The PX-EM algorithm for fast stable fitting of Henderson's mixed model. Genetics, Selection Evolution 32, 143-163 .
van Dyk, D. A. (2000). The nested EM algorithm. Statistica Sinica, 10,, 203-225 . abstract.
van Dyk, D. A. (2000). Fitting mixed-effects models using efficient EM-type algorithms. The Journal of Computational and Graphical Statistics, 9, 78-98. abstract.
Meng, X. L., van Dyk, D. A. (1999). Seeking efficient data augmentation schemes via conditional and marginal augmentation. Biometrika, 86 , 301-320, abstract.
Meng, X. L., van Dyk, D. A. (1998). Fast EM implementations for mixed-effects models. J. R. Statist. Soc., Ser. B., 60, 559-578, abstract.
Meng, X. L., van Dyk, D. A. (1997). The EM algorithm -- an old folk song sung to a fast new tune. Reading paper in J. R. Statist. Soc., Ser. B, 59, 511-567, abstract.
van Dyk, D. A., Meng, X. L. (1997). Some findings on the orderings and groupings of conditional maximizations within ECM-type algorithms. The Journal of Computational and Graphical Statistics, 6, 202-223, abstract.
van Dyk, D. A., Meng X. L., Rubin, D. B. (1995). Maximum likelihood estimation via the ECM algorithm: computing the asymptotic variance. Statistica Sinica 5, 55-75, abstract.

Refereed Publications in Edited Volumes and Encyclopedias

van Dyk, D. A., Hans, C. M. (2002). Accounting for Absorption Lines in Images Obtained with the Chandra X-ray Observatory, in Spatial Cluster Modelling (Editors: D. Denison and A. Lawson), CRC Press, London, 175-198.
van Dyk, D. A. (2002). Hierarchical models, data augmentation, and MCMC, in Statistical Challenges in Modern Astronomy III (Eds. G. J. Babu and E. D. Feigelson), Springer, New York, 41-56.
Meng, X. L. and van Dyk, D. A. (2000). The EM algorithm, invited entry,in The Encyclopedia of Mathematics, Kluwer Academic Publishers, Dordrecht.

Contributed and Invited Discussions

van Dyk, D. A. (2002). Invited comment on Setting Confidence Intervals on Bounded Parameters by M. Mandelkern. Statistical Science, 17, 149-172.
Zaslavsky, A. and van Dyk, D. A. (2002). Comment on A Note on the Estimation of the Multinomial Logit Model with Random Effects by Z. Chen and L. Kuo. The American Statistician, 56, 80-81.
van Dyk, D. A. (2002). Comment on Bayesian Adaptive Exploration by T. J. Loredo and D. F. Chernof. in, Statistical Challenges in Modern Astronomy III (Eds. G. J. Babu and E. D. Feigelson), Springer, New York, 70.
van Dyk, D. A. (2002). Comment on Statistical and astronomical problems arising in the Sloan Digital Sky Survey by M. Strauss. in, Statistical Challenges in Modern Astronomy III (Eds. G. J. Babu and E. D. Feigelson), Springer, New York, 124-126.
van Dyk, D. A. (1999). Comment on Quantifying surprise in the data and model verification by M. J. Bayarri and J. O. Berger. Bayesian Statistics VI, Oxford University Press, Oxford, 53-82, download PostScript file.
van Dyk, D. A. (1999). Comment on Simulated sintering: Markov chain Monte Carlo with spaces of varying dimension by J. S. Liu and C. Sabatti. Bayesian Statistics VI, Oxford University Press, Oxford, 389--413, download PostScript file.

Conference Proceedings and Other Publications

Hans, C. and van Dyk, D. A. (2002). Accounting for absorption lines in high energy spectra, in Statistical Challenges in Modern Astronomy III (Eds. G. J. Babu and E. D. Feigelson), Springer, New York, 429-430.
Sourlas, E., van Dyk, D. A., Kashyap, V. L., Drake, J. J. and Pease, D. (2002). Bayesian spectral analysis and metal abundance deficient stars, in Statistical Challenges in Modern Astronomy III (Eds. G. J. Babu and E. D. Feigelson), Springer, New York, 489-490.
Kang, H., van Dyk, D. A., Yu, Y., Connors, A., Kashyap, V. L., and Siemiginowska, A. (2002). Bayesian methods for addressing pile up in the Chandra X-ray observatory, in Statistical Challenges in Modern Astronomy III (Eds. G. J. Babu and E. D. Feigelson), Springer, New York, 449-450.
van Dyk, D. A., Meng, X. L. (2000). Algorithms based on data augmentation: A graphical representation and comparison, in Computing Science and Statistics: Proceedings of the 31st Symposium on the Interface (Eds. K. Berk and M. Pourahmadi), 230-239. Interface Foundation of North America, Fairfax Station, VA, download PostScript file or PDF file.
Meng, X. L., van Dyk, D. A. (1996). Minimum information ratio and relative augmentation function The Proceedings of the Statistical Computing Section of the American Statistical Association, 73-78, abstract.
Meng, X. L., van Dyk, D. A. (1995). Augmenting data wisely to speed up the EM algorithm. The Proceedings of the Statistical Computing Section of the American Statistical Association, 160-165, abstract.
van Dyk, D. A. (1995). Construction, Implementation, and Theory of Algorithms Based on Data Augmentation and Model Reduction. Ph.D. Thesis, University of Chicago, Department of Statistics, abstract.
van Dyk, D. A., Meng, X. L. (1994). Permuting CM steps within the ECM Algorithm. The Proceedings of the Statistical Computing Section of the American Statistical Association, 130-135, abstract.
van Dyk, D. A. (1993). Fitting log-linear models to contingency tables with incomplete observations. Technical Report No. 381, Department of Statistics, University of Chicago, abstract.

David van Dyk's Home Page

Research Links:

Teaching Links:

Publications and Manuscripts

Some Links