Partially identifying competing risks models: An application to the war on cancer

Abstract Competing risks models for discretely measured durations are partially identifying due to the unknown dependence structure between risks and the discrete nature of the outcome. This article develops a highly tractable bounds approach for underlying distributions of latent durations by exploiting the discreteness. Bounds are obtained from a system of nonlinear (in)equalities. I devise a sequential solution method that requires much less computational burden than existing methods. Asymptotic properties of bound estimators and a simple bootstrap procedure are provided. I apply the proposed approach to re-evaluate trends in cancer mortality extending the data studied in Honore and Lleras-Muney (2006). Estimated patterns differ from the original findings.

[1]  Charles F. Manski,et al.  Confidence Intervals for Partially Identified Parameters , 2003 .

[2]  Donald W. K. Andrews,et al.  Higher‐Order Improvements of a Computationally Attractive k‐Step Bootstrap for Extremum Estimators , 2002 .

[3]  J. Horowitz,et al.  Nonparametric Analysis of Randomized Experiments with Missing Covariate and Outcome Data , 2000 .

[4]  Xuelin Huang,et al.  Regression Survival Analysis with an Assumed Copula for Dependent Censoring: A Sensitivity Analysis Approach , 2008, Biometrics.

[5]  Ruixuan Liu A competing risks model with time‐varying heterogeneity and simultaneous failure , 2020 .

[6]  J. Heckman,et al.  New Methods for Analyzing Structural Models of Labor Force Dynamics , 1982 .

[7]  Azeem M. Shaikh,et al.  Inference for the identified set in partially identified econometric models , 2006 .

[8]  D. Andrews,et al.  Invalidity of the Bootstrap and The M Out Of N Bootstrap for Confidence Interval Endpoints Defined by Moment Inequalities , 2009 .

[9]  Federico A. Bugni Bootstrap Inference in Partially Identified Models Defined by Moment Inequalities: Coverage of the Identified Set , 2010 .

[10]  Adam M. Rosen,et al.  Confidence Sets for Partially Identified Parameters that Satisfy a Finite Number of Moment Inequalities , 2006 .

[11]  A. V. Peterson,et al.  Bounds for a joint distribution function with fixed sub-distribution functions: Application to competing risks. , 1976, Proceedings of the National Academy of Sciences of the United States of America.

[12]  Patrik Guggenberger,et al.  VALIDITY OF SUBSAMPLING AND “PLUG-IN ASYMPTOTIC” INFERENCE FOR PARAMETERS DEFINED BY MOMENT INEQUALITIES , 2007, Econometric Theory.

[13]  A. Blaes,et al.  Cardio-oncology Related to Heart Failure: Common Risk Factors Between Cancer and Cardiovascular Disease. , 2017, Heart failure clinics.

[14]  Ralf A. Wilke,et al.  A copula model for dependent competing risks , 2009 .

[15]  Gerard J. van den Berg,et al.  The identifiability of the mixed proportional hazards competing risks model , 2003 .

[16]  Jacques F. Carriere,et al.  Removing Cancer when it is Correlated with other Causes of Death , 1995 .

[17]  G. Logroscino,et al.  Incidence of cardiovascular disease and cancer in advanced age: prospective cohort study , 2008, BMJ : British Medical Journal.

[18]  V. Chernozhukov,et al.  Estimation and Confidence Regions for Parameter Sets in Econometric Models , 2007 .

[19]  J. Moslehi,et al.  Cardiovascular Toxic Effects of Targeted Cancer Therapies. , 2016, The New England journal of medicine.

[20]  John P. Klein,et al.  Estimates of marginal survival for dependent competing risks based on an assumed copula , 1995 .

[21]  Robert G. Valletta,et al.  The Effect of Extended Unemployment Insurance Benefits: Evidence from the 2012-2013 Phase-Out , 2015, SSRN Electronic Journal.

[22]  David C. Wheelock,et al.  Why do Banks Disappear? The Determinants of U.S. Bank Failures and Acquisitions , 2000, Review of Economics and Statistics.

[23]  H. Joe Multivariate Models and Multivariate Dependence Concepts , 1997 .

[24]  Do Extended Unemployment Benefits Lengthen Unemployment Spells? Evidence from Recent Cycles in the U.S. Labor Market , 2015 .

[25]  S. Bond,et al.  Bounds on the covariate-time transformation for competing-risks survival analysis , 2006, Lifetime data analysis.

[26]  Geert Ridder,et al.  Partial identification of the treatment effect distribution and its functionals , 2019, Journal of Econometrics.

[27]  David M Cutler,et al.  Are we finally winning the war on cancer? , 2008, The journal of economic perspectives : a journal of the American Economic Association.

[28]  Francesca Molinari,et al.  Microeconometrics with Partial Identification , 2020, 2004.11751.

[29]  Martin T. Wells,et al.  A Martingale Approach to the Copula-Graphic Estimator for the Survival Function under Dependent Censoring , 2001 .

[30]  Robert G. Valletta,et al.  Do Extended Unemployment Benefits Lengthen Unemployment Spells? , 2013, The Journal of Human Resources.

[31]  Daniel Krewski,et al.  Lung Cancer and Cardiovascular Disease Mortality Associated with Ambient Air Pollution and Cigarette Smoke: Shape of the Exposure–Response Relationships , 2011, Environmental health perspectives.

[32]  James J. Heckman,et al.  The identifiability of the competing risks model , 1989 .

[33]  Identification of a competing risks model with unknown transformations of latent failure times , 2005 .

[34]  D. Schoormans,et al.  Visualization of Cancer and Cardiovascular Disease Co-Occurrence With Network Methods. , 2017, JCO clinical cancer informatics.

[35]  Inference for Parameters Defined by Moment Inequalities: A Recommended Moment Selection Procedure , 2012 .

[36]  Bjørn Lomborg,et al.  The Bjørn-again economist The Skeptical Environmentalist. Measuring The Real State of The World , 2001 .

[37]  Brian P. McCall,et al.  Unemployment Insurance Rules, Joblessness, and Part-Time Work , 1996 .

[38]  Elie Tamer,et al.  Bayesian inference in a class of partially identified models , 2016 .

[39]  R. Nelsen An Introduction to Copulas , 1998 .

[40]  M. Crowder On assessing independence of Competing Risks when failure times are discrete , 1996, Lifetime data analysis.

[41]  Yanqin Fan,et al.  Partial identification and inference in censored quantile regression , 2018, Journal of Econometrics.

[42]  A. Blaes,et al.  Shared Risk Factors in Cardiovascular Disease and Cancer , 2016, Circulation.

[43]  H. Adami,et al.  Suicide and cardiovascular death after a cancer diagnosis. , 2012, The New England journal of medicine.

[44]  NONPARAMETRIC IDENTIFICATION OF ACCELERATED FAILURE TIME COMPETING RISKS MODELS , 2013, Econometric Theory.

[45]  Jörg Stoye,et al.  More on Confidence Intervals for Partially Identified Parameters , 2008 .

[46]  Ann Berrington,et al.  Marriage or cohabitation: a competing risks analysis of first‐partnership formation among the 1958 British birth cohort , 2000 .

[47]  Alison L. Booth,et al.  The Hazards of Doing a PhD: An Analysis of Completion and Withdrawal Rates of British PhDs in the 1980s , 1995 .

[48]  John M. Quigley,et al.  Mortgage Terminations, Heterogeneity and the Exercise of Mortgage Options , 1999 .

[49]  Yi-Hau Chen Semiparametric marginal regression analysis for dependent competing risks under an assumed copula , 2010 .

[50]  The Recall and New Job Search of Laid-Off Workers: A Bivariate Proportional Hazard Model with Unobserved Heterogeneity , 2007, The Review of Economics and Statistics.

[51]  J. Hausman,et al.  Flexible parametric estimation of duration and competing risk models , 1990 .

[52]  Glenn T. Sueyoshi Semiparametric proportional hazards estimation of competing risks models with time-varying covariates , 1992 .

[53]  Andrew Chesher,et al.  Generalized instrumental variable models , 2017 .

[54]  Francesca Molinari Microeconometrics with partial identification , 2020, Handbook of Econometrics.

[55]  Ian Gregory�?Smith Fired or Retired? A Competing Risks Analysis of Chief Executive Turnover , 2009 .

[56]  Bruce D. Meyer,et al.  The impact of the potential duration of unemployment benefits on the duration of unemployment , 1990 .

[57]  Simon M. S. Lo,et al.  IDENTIFIABILITY OF THE SIGN OF COVARIATE EFFECTS IN THE COMPETING RISKS MODEL , 2017, Econometric Theory.

[58]  Roger Newson,et al.  Parameters behind “Nonparametric” Statistics: Kendall's tau, Somers’ D and Median Differences , 2002 .