Inference for the complier-average causal effect from longitudinal data subject to noncompliance and missing data, with application to a job training assessment for the unemployed.(Statistical Data Included)

Linda H.Y Yau is Senior Statistician, GlaxoSmithKline, Research Triangle Park, NC 27709 (E-mail: 1y20993@gskcom). Roderick J. Little is Professor, Department of Biostatistics, School of Public Health, University of Michigan, Ann Arbor, MI 48109 (E-mail: [email protected]). This work was supported by NIMH grant P30 MH38330-12 and was completed in part while the first author was graduate student, Department of Biostatistics, School of Public Health, University of Michigan, Ann Arbor, MI 48019. The authors thank Rick Price, Amiram Vinokur for their valuable support and suggestions, and Don Rubin for his helpful advice.