Addressing Endogeneity in count data using a two-stage copula generated regressor approach

This research project addresses a common problem in studying medical data: understanding the true effects of different treatments when the data is not from a controlled experiment. Specifically, we are looking at patients with systemic lupus erythematosus (SLE) and the number of infections they get. Often, traditional methods need special, hard-to-find data to ensure accurate results. Our new method skips this requirement by using advanced statistical techniques to account for unmeasured factors affecting the results. By applying this method to SLE patient data, we aim to see how Glucocorticoid influences infection rates. This research could lead to better treatment strategies, improving patient care. The partner organization will benefit from a practical and widely applicable tool to analyze complex medical data more accurately, even when ideal data conditions are not met.

Faculty Supervisor:

Hui Xie

Student:

Partner:

Arthritis Research Canada

Discipline:

Mathematics

Sector:

Professional, scientific and technical services

University:

Simon Fraser University

Program: