Incorporating Auxiliary Information by Joint Modeling of Pseudo Data and Length Biased Data

• Activity Number: 414 - Advances in Estimation Methods
• Type: Contributed
• Date/Time: Tuesday, July 31, 2018 : 2:00 PM to 3:50 PM
• Sponsor: SSC
• Author(s): Yidan Shi* and Leilei Zeng and Mary E. Thompson and Suzanne Tyas
• Companies: University of Waterloo and University of Waterloo and University of Waterloo and University of Waterloo
• Keywords: survival analysis; left-truncation; combined cohorts; auxiliary information

Abstract

Time-to-event data are often limited in sample size, subject to length-biased sampling, or limited to a specic time/age range. Most of the existing methods are based solely on the information contained in the objective sample. In some situations, information for a sample from the same or similar cohorts, having more observations or a longer observation time, may be obtained from literature. However, we may not be able to access the source data, or may have trouble combining the two samples. In this paper, we introduce a way to incorporate the information from the auxiliary sample by creating pseudo datasets which match the form and conditions of the observed sample, eliminating the length-biased sampling problem and improving eciency. The method is illustrated by both a real dataset and simulation studies.