Bayesian Estimation of the Complier Average Casual Effect

Martijn van Hasselt, Timothy Ferland, Jeremy Bray and Arnie Aldridge
Additional contact information
Timothy Ferland: University of North Carolina at Greensboro, Department of Economics
Arnie Aldridge: RTI International

No 17-14, UNCG Economics Working Papers from University of North Carolina at Greensboro, Department of Economics

Abstract: We analyze two cases of randomized experiments with noncompliance. In the first case, individuals in the control group do not have access to the treatment and non- compliance only occurs in the treatment group. In the second case, which is com- mon in clinical studies, individuals in the control group are given a placebo. In this case, noncompliance occurs in both the treatment and control group. We present a Bayesian method-of-moments approach for estimating the complier average causal ef- fect (CACE). This procedure is valid under weak exclusion restrictions. This approach is contrasted with Gibbs sampling and data augmentation in a parametric model. The various procedures are evaluated in a simulation experiment and applied to clinical trial data from the COMBINE study.

Keywords: Noncompliance; casual effects; methods of moments; principal stratification (search for similar items in EconPapers)
JEL-codes: C11 C14 C21 C26 (search for similar items in EconPapers)
Pages: 33 pages
Date: 2017-12-21
New Economics Papers: this item is included in nep-ecm and nep-exp
References: View references in EconPapers View complete reference list from CitEc
Citations:

Downloads: (external link)
https://bryan.uncg.edu/wp-content/uploads/2018/02/17-14-Bayesian-Estimation.pdf Full text (application/pdf)
Our link check indicates that this URL is bad, the error code is: 404 Not Found

Related works:
This item may be available elsewhere in EconPapers: Search for items with the same title.

Export reference: BibTeX RIS (EndNote, ProCite, RefMan) HTML/Text

Persistent link: https://EconPapers.repec.org/RePEc:ris:uncgec:2017_014

Access Statistics for this paper

More papers in UNCG Economics Working Papers from University of North Carolina at Greensboro, Department of Economics UNC Greensboro, Department of Economics, PO Box 26170, Bryan Building 462, Greensboro, NC 27402. Contact information at EDIRC.
Bibliographic data for series maintained by Albert Link ().