Biased Sampling, Left Truncation and Survival Analysis, Lecture II - Dr. Jing Qin
Biostatistics Branch Seminar Series
September 17, 2019 | 10:30 AM – 12:00 PM
NCI Shady Grove 7WEST032/034 Rockville, MD
Jing Qin, Ph.D.,
Mathematical Statistician, Biostatistics Research Branch
National Institute of Allergy & Infectious Diseases, NIH
Biased sampling occurs when a proper randomization cannot be achieved; in this case, the observed sample will not be representative of the population of interest. Biased sampling problems appear in many areas of research, including, medicine, epidemiology and public health, social sciences, and economics. Left truncation and length-biased data are clearly encountered in applications of renewal processes, etiologic studies, genome-wide linkage studies, epidemiologic cohort studies, cancer prevention trials, and studies of labor economy. In observational studies, a prevalent cohort cohort design that draws samples from individuals with a condition or disease at the time of enrollment is generally more efficient and practical. The recruited patients who have already experienced an initiating event are followed prospectively for the failure event (e.g. disease progression or death) or are right censored. Under this sampling design, individuals with longer survival times measured from the onset of the disease are more likely to be included in the cohort, whereas those with shorter survival times are unconsciously excluded. Finding appropriate adjustments for the potential selection bias in analyzing length-biased data or more general biased sampling problems has been a long standing statistical problem.
In lecture II, I will present some of the latest results on analyzing length biased survival time data, including Vardi's multiplicative censoring problem, and inferences based on the Cox regression model with length biased survival data. If time permits, I will also discuss the well known pool adjacent violators algorithm and its combination with the EM what does this stand for? algorithm together for estimating shape constrained inference, including estimation of monotonic decreasing density, cumulation hazard, or distribution function based on current status data, etc.
**The mission of the Biostatistics Branch (BB) is to be an outstanding biostatistics unit that can contribute to the understanding of cancer etiology and to improve public health by the development and application of quantitative methods. The BB Investigators develop statistical methods and data resources to strengthen observational studies, intervention trials, and laboratory investigations of cancer.**