Measuring the Impact of Nonignorability in Panel Data with Non-Monotone Nonresponse

SUMMARY The analysis of panel data with non-monotone nonresponse often relies on the critical and untestable assumption of ignorable missingness. It is important to assess the consequences of departures from the ignorability assumption. Non-monotone nonresponse, however, can often make such sensitivity analysis infeasible because the likelihood functions for alternative models involve high-dimensional and difficult-to-evaluate integrals with respect to missing outcomes. We develop an extension of the local sensitivity method that overcomes computational difficulty and completely avoids fitting alternative models and evaluating these high-dimensional integrals. The proposed method is applicable to a wide range of panel outcomes. We apply the method to a Smoking Trend dataset where we relax the standard ignorability assumption and evaluate how smoking-trend estimates in different groups of US young adults are affected by alternative assumptions about the missing-data mechanism. The main finding is that the standard estimate in the black male group is sensitive to nonignorable missingness but those in other groups are reasonably robust. Copyright © 2010 John Wiley & Sons, Ltd.

[1]  A. Troxel,et al.  AN INDEX OF LOCAL SENSITIVITY TO NONIGNORABILITY , 2004 .

[2]  D. Rubin,et al.  Ignorability and Coarse Data , 1991 .

[3]  P S Albert,et al.  A Transitional Model for Longitudinal Binary Data Subject to Nonignorable Missing Data , 2000, Biometrics.

[4]  M. Kenward.  Selection models for repeated measurements with non-random dropout: an illustration of sensitivity. , 1998, Statistics in medicine.

[5]  Daniel F Heitjan,et al.  Sensitivity analysis of causal inference in a clinical trial subject to crossover , 2004, Clinical trials.

[6]  Roderick J. A. Little,et al.  Adjusting for Nonignorable Drop-Out Using Semiparametric Nonresponse Models: Comment , 1999 .

[7]  Cheti Nicoletti,et al.  Nonresponse in dynamic panel data models , 2006 .

[8]  M. Kenward,et al.  Informative Drop‐Out in Longitudinal Data Analysis , 1994 .

[9]  G Molenberghs,et al.  Sensitivity Analysis for Nonrandom Dropout: A Local Influence Approach , 2001, Biometrics.

[10]  John S. Preisser,et al.  Analysis of Smoking Trends with Incomplete Longitudinal Binary Responses , 2000 .

[11]  D. Follmann,et al.  An approximate generalized linear model with random effects for informative missing data. , 1995, Biometrics.

[12]  Michel Chavance,et al.  Sensitivity analysis of longitudinal binary data with non-monotone missing values. , 2004, Biostatistics.

[13]  J. Tobin.  Estimation of Relationships for Limited Dependent Variables , 1958 .

[14]  Jerry A. Hausman,et al.  Attrition Bias in Experimental and Panel Data: The Gary Income Maintenance Experiment , 1979 .

[15]  P. Diggle,et al.  Analysis of Longitudinal Data , 2003 .

[16]  Daniel F Heitjan,et al.  An index of local sensitivity to nonignorable drop‐out in longitudinal modelling , 2005, Statistics in medicine.

[17]  Hui Xie,et al.  A local sensitivity analysis approach to longitudinal non‐Gaussian data with non‐ignorable dropout , 2008, Statistics in medicine.

[18]  J. Copas,et al.  Local sensitivity approximations for selectivity bias , 2001 .

[19]  Donald B. Rubin,et al.  Combining Panel Data Sets with Attrition and Refreshment Samples , 1998 .

[20]  James M. Robins,et al.  Semiparametric Regression for Repeated Outcomes With Nonignorable Nonresponse , 1998 .

[21]  Stuart R. Lipsitz,et al.  Analysis of longitudinal data with non‐ignorable non‐monotone missing values , 2002 .

[22]  Smoking cessation during previous year among adults--United States, 1990 and 1991. , 1993, MMWR. Morbidity and mortality weekly report.

[23]  Jasjeet S. Sekhon,et al.  Genetic Optimization Using Derivatives , 2011, Political Analysis.

[24]  James J. Heckman,et al.  Choosing Among Alternative Nonexperimental Methods for Estimating the Impact of Social Programs: the Case of Manpower Training , 1989 .

[25]  J. Horowitz,et al.  Identification and estimation of statistical functionals using incomplete data , 2006 .

[26]  J. Robins,et al.  Adjusting for Nonignorable Drop-Out Using Semiparametric Nonresponse Models , 1999 .

[27]  M D Schluchter,et al.  Methods for the analysis of informatively censored longitudinal data. , 1992, Statistics in medicine.

[28]  J. Copas,et al.  Inference for Non‐random Samples , 1997 .

[29]  Geert Molenberghs,et al.  A Local Influence Approach Applied to Binary Data from a Psychiatric Study , 2003, Biometrics.

[30]  Daniel F Heitjan,et al.  A Simple Local Sensitivity Analysis Tool for Nonignorable Coarsening: Application to Dependent Censoring , 2006, Biometrics.

[31]  D. O. Scharfstein.  Adjusting for nonignorable dropout using semiparametric nonresponse models (with discussion) , 1999 .

[32]  K. Judd.  Numerical methods in economics , 1998 .

[33]  Roderick J. A. Little,et al.  Modeling the Drop-Out Mechanism in Repeated-Measures Studies , 1995 .

[34]  M. Kenward,et al.  Informative dropout in longitudinal data analysis (with discussion) , 1994 .

[35]  P. Albert,et al.  Models for longitudinal data: a generalized estimating equation approach. , 1988, Biometrics.

[36]  Richard A. Ashley,et al.  Assessing the credibility of instrumental variables inference with imperfect instruments via sensitivity analysis , 2009 .

[37]  Jeffrey M. Woodbridge.  Econometric Analysis of Cross Section and Panel Data , 2002 .

[38]  Yi Qian,et al.  Do National Patent Laws Stimulate Domestic Innovation in a Global Patenting Environment? A Cross-Country Analysis of Pharmaceutical Patent Protection, 19782002 , 2007, The Review of Economics and Statistics.

[39]  Daniel F Heitjan,et al.  Impact of nonignorable coarsening on Bayesian inference. , 2006, Biostatistics.

[40]  D. Rubin.  INFERENCE AND MISSING DATA , 1975 .

[41]  W Vach,et al.  Logistic regression with incompletely observed categorical covariates--investigating the sensitivity against violation of the missing at random assumption. , 1995, Statistics in medicine.