Content area

Abstract

Control versus treatment clinical trials are prone to averaging over individual effects to produce a one size fits all conclusion of whether a drug/procedure have efficacy, on average. However, it is often the case that subjects have their own unique response to treatment that are not necessarily captured by an average effect size. We have developed an infinite mixture model that can produce a nonparametric estimate of the latent distribution of treatment effects for a population. The estimated distribution of treatment effects enables researchers to make conclusions about what percentage of the population will have a treatment effect of a specified size (i.e. medically significant), without assuming a parametric form on their data. This development opens the door for personalized medicine to be statistically modeled in two arm clinical trials.

Details

1010268
Title
Nonparametric Estimation of Latent Treatment Effects for Two Armed Clinical Trials Using Infinite Mixture Models
Number of pages
212
Publication year
2025
Degree date
2025
School code
0032
Source
DAI-B 87/1(E), Dissertation Abstracts International
ISBN
9798288852756
Committee member
Kurum, Esra; Yao, Weixin
University/institution
University of California, Riverside
Department
Applied Statistics
University location
United States -- California
Degree
Ph.D.
Source type
Dissertation or Thesis
Language
English
Document type
Dissertation/Thesis
Dissertation/thesis number
32045467
ProQuest document ID
3231748631
Document URL
https://www.proquest.com/dissertations-theses/nonparametric-estimation-latent-treatment-effects/docview/3231748631/se-2?accountid=208611
Copyright
Database copyright ProQuest LLC; ProQuest does not claim copyright in the individual underlying works.
Database
ProQuest One Academic