Jianping Sun

Contact Information:
Address:	Department of Statistics 326 Thomas Building The Pennsylvania State University University Park, PA, 16802
Email:	jxs1021@psu.edu
Phone:	814-865-8635
Homepage:	http://www.stat.psu.edu/~jxs1021
CV

Education:

Aug. 2005 - Now:	PhD Candidate in Statistics, Department of Statistics, Penn State University, U.S.A. Advisor: Bruce G. Lindsay
Aug. 2002 - Jun. 2005:	M.S. in Statistics, Department of Statistics, Nankai University, China. Advisor: Runchu Zhang
Aug. 1998 - Jun. 2002:	B.S. in Statistics, Department of Statistics, Nankai University, China.

Research:

My current research focuses on the composite likelihood method and its application to long sequence data.

Maximum likelihood is a popular statistical method largely because it provides estimators with optimal statistical efficiency. However, many realistic statistical models are so complex in structure that it becomes computationally infeasible to find the MLE, especially in large data sets. One approach to solving this problem is the method of composite likelihood, which is constructed by taking a product of likelihood terms, each one of which is a marginal or conditional sublikelihood. Composite likelihood can reduce the complexity of computation at the price of some loss of efficiency. The focus of my research has been on methods for constructing a composite likelihood so that it maintains most efficiency and economical computation at the same time.

There are many applications for composite likelihood method in longitudinal data, survival analysis, spatial data, and genetic data. My application is the analysis of long sequence data generated in biology, such as SNP data. The research question is to estimate the ancestral distribution from the observed descendant sequences, considering realistic genetic complexities such as mutation and recombination. We have developed a statistical model to estimate the ancestral distribution. However, there is an enormous computation challenge when applying it on data due to an enormous number of recombination possibilities. Therefore, we apply composite likelihood as an approximation to solve the problem.

Publication and Manuscripts:

B. Lindsay, G. Yi, and J. Sun. Issues and Strategies in the Selection of Composite Likelihoods. Submitted.

Teaching:

Fall 2009	Instructor of Stat/Math 414: Probability Theory - Syllabus
Summer 2009	Instructor of Stat 200 Elementary Statistics - Syllabus
Summer 2008	Instructor of Stat 401: Experimental Methods - Syllabus

Conferences and Presentations:

Aug.1 - 6, 2009	2009 Joint Statistical Meeting, Washington D.C. Contributed talk: Composite likelihood in Hayplotype sequence data.
May 22, 2009	The 2009 Rao Prize Conference , State College, Pennsylvania Poster presentation: Composite likelihood: Issues in Efficiency.
Mar. 14 -18, 2009	ENAR 2009, San Antonio, Texas Contributed talk: Composite likelihood: Issues in Efficiency.

Other Professional Activities:

Spring and Fall 2007: Student Consultant at Statistical Consulting Center in the Statistics Department, Penn State University.

Useful Links:

Journals and Research Resources			Societies, Institutes, and Associations
	Current Index to Statistics (CIS) JSTOR MathSciNet AMS Mathematical Reviews on the Web Web of Science link to Science Citation Index Penn State Library			American Statistical Association (ASA) ENAR (Eastern North American Region of IBS) Institute of Mathematical Statistics (IMS) The Royal Statistical Society Statistical Society of Canada
Statistical Computing			Others (non-academic)
		R statistical sofware High Performance Computing		Local new and entertainment The Daily Collegain Food network cooking recipes

Latest revision: Nov.15, 2009