Use of Bayesian Non-Parametrics for Problems in Causal Inference - Dr. Mike Daniels
DCEG Events
May 29, 2019 | 10:30 AM – 11:30 AM
NCI Shady Grove Rockville, MD
BIOSTATISTICS BRANCH SEMINAR SERIES PRESENTS
Speaker
Mike Daniels, Ph.D., Professor,
Department of Statistics & Andrew Banks Family Endowed Chair,
University of Florida
Abstract
The authors propose a general Bayesian nonparametric (BNP) approach to causal inference in the point treatment setting. The joint distribution of the observed data (outcome, treatment, and confounders) is modeled using an enriched Dirichlet process. The combination of the observed data model and causal assumptions allows us to identify any type of causal effect - differences, ratios, or quantile effects, either marginally or for subpopulations of interest. The proposed BNP model is well-suited for causal inference problems, as it does not require parametric assumptions about the distribution of confounders and naturally leads to a computationally efficient Gibbs sampling algorithm. By flexibly modeling the joint distribution, the authors are also able to impute (via data augmentation) values for missing covariates within the algorithm under an assumption of ignorable missingness, obviating the need to create separate imputed data sets. This approach for imputing the missing covariates has the additional advantage of guaranteeing congeniality between the imputation model and the analysis model, and because the authors use a BNP approach, parametric models are avoided for imputation. The performance of the method is assessed using simulation studies. The method is applied to data from a cohort study of human immunodeficiency virus/hepatitis C virus co-infected patients. If time permits, the authors discuss extensions to dynamic treatments and time-varying confounders and causal mediation with multiple mediators, and a new approach for confounder (and mediator) selection.
Joint work with Jason Roy (Rutgers) and Kumaresh Dhara (UF)
**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.**