非常抱歉，
你要访问的页面不存在，
非常抱歉，
你要访问的页面不存在，
非常抱歉，
你要访问的页面不存在，
验证码：
职称:professor
所属学校:Carnegie Mellon University
所属院系:Department of Statistics
所属专业:Mathematical Statistics and Probability
联系方式:(412) 268-2717
Why be a statistician? For me the answer is simple - it's fun! As a statistician, I have had the opportunity to meet investigators from many different disciplines, learn about their research, and contribute my skills and knowledge to help them better understand and solve their problems. Whether it be studying changes in infant mortality rates in nineteenth century Sweden, biological correlates of depression, the ethics of randomized controlled clinical trials, or the relative efficacy of coronary bypass surgery to coronary angioplasty, every statistical collaboration is an opportunity to learn new things and to contribute to the advancement of knowledge in a field. Part of my fascination with statistics is the dynamic interplay between the theory and the practice of statistics as they come together to solve real-world problems. Work with Rob Kass, Ruey Tsay, and Teresa Lam, then a graduate student, on nonlinear models with time series errors, which provided a new statistical method for the analysis of biological rhythm data, grew out of my collaboration with Dr. David Jarrett, a psychiatrist at the University of Pittsburgh. A project I worked on with Satish Iyengar was in the area of combining information from independent studies, which is known as meta-analysis. We considered the "file-drawer" problem, which is that researchers often do not publish their results when they are not significant (and instead "put them in a file drawer"). This work grew out of our participation in a meta-analysis of the efficacy of treatment for aphasia with Drs. Fromm and Holland, speech pathologists at the University of Pittsburgh. An essential component in the training of statisticians is hands on participation in research projects. I have taken a special pleasure in this kind of interaction with students. Some recent examples include work with Paul Gustafson on the analysis of bivariate data from multi-center cancer clinical trials, Nancy Paul on the cost-effectiveness of treatment for depression, and Donna Pauler on applications of mixture models in the analysis of schizophrenia data.
Why be a statistician? For me the answer is simple - it's fun! As a statistician, I have had the opportunity to meet investigators from many different disciplines, learn about their research, and contribute my skills and knowledge to help them better understand and solve their problems. Whether it be studying changes in infant mortality rates in nineteenth century Sweden, biological correlates of depression, the ethics of randomized controlled clinical trials, or the relative efficacy of coronary bypass surgery to coronary angioplasty, every statistical collaboration is an opportunity to learn new things and to contribute to the advancement of knowledge in a field. Part of my fascination with statistics is the dynamic interplay between the theory and the practice of statistics as they come together to solve real-world problems. Work with Rob Kass, Ruey Tsay, and Teresa Lam, then a graduate student, on nonlinear models with time series errors, which provided a new statistical method for the analysis of biological rhythm data, grew out of my collaboration with Dr. David Jarrett, a psychiatrist at the University of Pittsburgh. A project I worked on with Satish Iyengar was in the area of combining information from independent studies, which is known as meta-analysis. We considered the "file-drawer" problem, which is that researchers often do not publish their results when they are not significant (and instead "put them in a file drawer"). This work grew out of our participation in a meta-analysis of the efficacy of treatment for aphasia with Drs. Fromm and Holland, speech pathologists at the University of Pittsburgh. An essential component in the training of statisticians is hands on participation in research projects. I have taken a special pleasure in this kind of interaction with students. Some recent examples include work with Paul Gustafson on the analysis of bivariate data from multi-center cancer clinical trials, Nancy Paul on the cost-effectiveness of treatment for depression, and Donna Pauler on applications of mixture models in the analysis of schizophrenia data.
