SSI 2017 New Course Spotlight - Survival Analysis with Bindu Viswanathan
Bindu Viswanathan received her Ph.D. in Biostatistics in 1999 from Emory University. For the next three years, She worked as a Biostatistician at Merck and Novartis, as the statistics lead on Phase III clinical trials in the fields of HIV-AIDS, anxiety/depression and eye diseases.
Missing teaching and the academic environment, she moved back to Emory as research faculty in 2002 to teach and work on collaborative research projects with the Emory schools of Nursing, Medicine, and Public Health, as well as the CDC and VA healthcare systems. Bindu moved to Austin in 2006, and went back to school to earn an M.S. in Conservation Biology from Texas State. She started teaching at UT in Spring 2012 and has been here ever since.
What is the main goal of this course?
The main goal of the course is to introduce the special constraints when working with failure-time (survival) data, and learn the methods of analysis specific to this type of data.
What is the principal scope of this course?
Participants in this course will learn how the methods of analysis for survival data differ from the usual methods for numerical data, learn the analysis methods specific to failure-time data, their interpretation, and applications. I will be giving examples using statistical software (R and possibly other software as well). So after this course, participants will be able to run their own analysis and interpret the results.
What makes you excited about teaching this course?
My doctoral dissertation was on the analysis of correlated failure-time data using frailty models. Most introductory statistics courses teach common methods for numerical and categorical data. So it is exciting to be able to offer a course that is specific to failure-time data, which students don’t get a chance to see usually.
How much background knowledge or experience in this subject is required to be able to follow the course material?
A course in introductory statistics and regression analysis will help. But I will try to always present examples, and break down the concepts in a logical and meaningful way so that students with just a cursory statistics background will be able to still get some value out of it.
What skills and knowledge can participants expect to acquire by the end of the course?
Participants who take this course will then be able to perform the analysis for failure-time data, be able to interpret the results, and recognize the constraints and possibilities with this type of data. I will be providing examples using statistical software (R, and possibly other software as well) so that participants will be able to run the analysis themselves after taking this course.
Last year you taught a brand new SSI course. Do you have any remarks about your experience?
Last year I taught Nonparametric Statistics. I had a full class, with participants ranging from one undergraduate student to several doctoral students, professionals, and faculty. Participants had varying degrees of familiarity with statistics, and different softwares they were comfortable with. It was challenging and fun to try to craft a course that everyone would find useful. From the feedback I got directly after the course from several participants, it appears that they enjoyed it and found it very useful. One person who had been taking SSI courses for several years told me this was one of the best he had taken.
How will teaching this new course be different? What are you looking forward to with Survival Analysis?
I usually conduct a survey on the first day, to gauge the expectations of every participant. While I have material prepared ahead of time, I adapt and expand the course as we go along, to match the interests and specific needs of participants in the class. So it is a very dynamic process for me. I am looking forward to this challenge, and to being able to cover ground on all the traditional methods in survival analysis.
The Department of Statistics and Data Sciences at The University of Texas at Austin is proud to host its 10th annual UT Summer Statistics Institute (SSI) May 22-25, 2017.