I-Shou Chang, Ph.D.
National Institute of Cancer Research
and Division of Biostatistics and Bioinformatics
Director, Office of Biomedical Informatics
EDUCATIONPh.D., Mathematics, Columbia University, New York, USA, 1976
M.A., Mathematics, Wayne State University, Detroit, MI, USA, 1972
B.S., Mathematics, National Tsing-Hua University, Taiwan, 1969
- Director, Center of Biomedical Databases, National Health Research Institutes, Taiwan (2007-present)
- Investigator, National Institute of Cancer Research and Division of Biostatistics and Bioinformatics, National Health Research Institutes, Taiwan (2006-present)
- Director, Department of Research Resources, National Health Research Institutes, Taiwan (2000-2007)
- Investigator, President’s Laboratory, National Health Research Institutes, Taiwan (2000-2006)
- Visiting Professor, Statistics Department, Columbia University, USA (1989-1990)
- Professor, Mathematics Department, National Central University, Taiwan (1982-present)
- Adjunct Research Fellow, Institute of Mathematics, Academia Sinica, Taiwan (1982-1992)
- Associate Professor, Mathematics Department, National Central University, Taiwan (1976-1982)
RESEARCH INTERESTSDr. Chang is interested in the developments of statistical models, inference methods, and computational algorithms that have applications in health sciences and other areas. These include sequential analysis, survival analysis, and molecular statistics, among others. Both Monte Carlo methods and asymptotic methods are employed to study models arisen from these studies. Specific models considered include semiparametric frailty models based on right-censored data or current status data, counting process models for event history data, Bayesian models using random Bernstein polynomials for shape-restricted inferences, etc.
RESEARCH ACTIVITIES & ACCOMPLISHMENTSHere are some recent results. A verifiable identifiability condition and an efficient computation algorithm are proposed for the profile likelihood theory for frailty models, which have wide applications in modern epidemiology. Several papers on Bernstein polynomials may have the potential to indicate a major alternative approach in the important area of shape-restricted inferences.
HONORS & AWARDS
- Outstanding Research Award, National Science Council, Taiwan, 1994-1996
- Excellence Research Award, National Science Council, Taiwan, l990-1993
- Elected Member, International Statistical Institutes, Taiwan, 1992
- Sun Yat-Sen Award for Outstanding Research, Taiwan, 1988
- Chang IS, HsiungCA, Wen CC, Wu YJ and Yang CC. Non-parametric maximum likelihood estimation in a semiparametric mixture model for competing risks data. Scandinavian J. Statistics, 34:870-895, 2007.
- Chang IS, Chien LC, Hsiung CA, Wen CC and Wu YJ. Shape restricted regression with random Bernstein polynomials. In R. Liu, W. Strawderman and C.H. Zhang (eds), Complex Dataset and Inverse Problems. IMS Lecture Notes-Monograph Series, 54:187-202, 2007.
- Chang IS, HsiungCA, Wu YJ and Yang CC. Bayesian survival analysis using Bernstein polynomials. Scandinavian J. Statistics, 32:447-466, 2005.
- Chang IS, Hsiung CA, Wang MC and Wen CC. An asymptotic theory for the non-parametric maximum likelihood estimate in the Cox-gene model. Bernoulli, 11:863-892, 2005.
- Chang IS and Hsiung CA. Asymptotic consistency of the maximum likelihood estimate in positron emission tomography and applications. The Annals of Statistics, 22:1871-1883, 1994.
- Chang IS and Hsiung CA. Information and asymptotic efficiency in generalized proportional hazards models for counting processes. The Annals of Statistics,22:1275-1298, 1994.
- Chang IS and Hsiung CA. Likelihood process in parametric model of censored data with staggered entry-asymptotic properties and applications. Journal of Multivariate Analysis, 24:31-45, 1988.