1. School of Mathematical Sciences, Peking University, Beijing 100871, China; 2. Beijing International Center for Mathematical Research, Peking University, Beijing 100871, China; 3. Department of Biostatistics, University of Washington, Seattle, WA 98195, USA; 4. Biostatistics Unit, HSR&D Center of Excellence, VA Puget Sound Health Care System, Seattle, WA 98101, USA
Principal strata are defined by the potential values of a posttreatment variable, and a principal effect is a causal effect within a principal stratum. Identifying the principal effect within every principal stratum is quite challenging. In this paper, we propose an approach for identifying principal effects on a binary outcome via a pre-treatment covariate. We prove the identifiability with single post-treatment intervention under the monotonicity assumption. Furthermore, we discuss the local identifiability with multicomponent intervention. Simulations are performed to evaluate our approach. We also apply it to a real data set from the Improving Mood-Promoting Access to Collaborate Treatment program.
. Identifiability of causal effects on a binary outcome within principal strata[J]. Frontiers of Mathematics in China, 2011, 6(6): 1249-1263.
Wei YAN, Peng DING, Zhi GENG, Xiaohua ZHOU. Identifiability of causal effects on a binary outcome within principal strata. Front Math Chin, 2011, 6(6): 1249-1263.
Everitt B S, Hand D J. Finite Mixture Distributions. London: Chapman and Hall, 1981
2
Frangakis C E, Rubin D B. Principal stratification in causal inference. Biometrics , 2002, 58: 21-29 doi: 10.1111/j.0006-341X.2002.00021.x
3
Gelman A, Carlin J B, Stern H S, Rubin D B. Bayesian Data Analysis. 2nd ed. London: Chapman and Hall/CRC, 2004
4
Goodman L A. Exploratory latent structure analysis using both identifiable and unidentifiable models. Biometrika , 1974, 61: 215-231 doi: 10.1093/biomet/61.2.215
5
Holland P. Statistics and causal inference. J Amer Statist Assoc , 1986, 81(396): 945-960 doi: 10.2307/2289064
6
Roche K B, Miglioretti D L, Zeger S L, Rathouz P J. Latent variable regression for multiple discrete outcomes. J Amer Statist Assoc , 1997, 92: 1375-1386 doi: 10.2307/2965407
7
Rubin D B. Comment on “Randomization analysis of experimental data: the Fisher randomization test” by D. Basu. J Amer Statist Assoc , 1980, 75: 591-593 doi: 10.2307/2287653
8
Rubin D B. Comments on “Statistics and causal inference” by Paul Holland: Which ifs have causal answers. J Amer Statist Assoc , 1986, 81: 961-962 doi: 10.2307/2289065
9
Rubin D B. Direct and indirect causal effects via potential outcomes. Scand J Stat , 2004, 31: 161-170 doi: 10.1111/j.1467-9469.2004.02-123.x
10
Zhang J L, Rubin D B, Mealli F. Likelihood-based analysis of causal effects via principal stratification: new approach to evaluating job-training programs. J Amer Statist Assoc , 2009, 104: 166-176 doi: 10.1198/jasa.2009.0012
