Seminar By Dr. Lianfen Qian
The School of Computer Science welcomes you to join us for a seminar by Dr. Lianfen Qian on Friday, April 10th, 2015 in Reynolds 219 at 2pm.
Detecting Change Points in Cure Models
My research interests are in time series analysis and regression modeling, survival analysis, environmental statistics, parameter and function estimation, weak convergence of stochastic processes and their applications to statistics. I collaborate with various researchers from different institutions and different fields including biology, computer science and the government.
In analyzing life time data, it has been observed that the hazard rate changes its pattern once the treatment shows its effect. It may change as well when the patient ages or recurrent event happens. This phenomenon is related to the multiple change point problems. For certain diseases, a substantial proportion of patients may now be cured by treatments. The patients who are cured are called immune or long-term survivors, whereas the remaining patients who develop a recurrence of the disease are called susceptibles. The population of interest may thus be divided into two groups and can be regarded as a mixture of these two types of patients. Those models are referred to cure models. Thus, cure models have become increasingly popular in the analysis of data from clinical trials due to treatment improvements. This talk discusses the recent developments in the multiple change point detection and estimation in hazard models when both long-term survivors and covariate effects are present for censored life time data.