Cox proportional hazards models with frailty for negatively correlated employment processes. (English) Zbl 1471.62224
Summary: In promotion discrimination cases, individuals affected by discrimination may decide to retire earlier than otherwise. Two Cox proportional hazards models are used to describe the promotion process from non-retired employees and the retirement process, respectively. To account for a potential negative correlation between the two outcomes, promotion and retirement, frailty terms are introduced. Model diagnoses in the presence of unobserved frailty terms are difficult. Therefore, the robustness of the parameter estimates to the fitting of an unnecessary frailty or a frailty distribution or form different from the underlying one is examined. The data from a reverse discrimination case, Alexander v. Milwaukee, are analyzed. The original finding of liability relying on a statistically significant coefficient for membership in the legally protected group (White-male) is shown to be robust to several choices of the frailty model. This provides further support for the court’s decision.
MSC:
| 62-08 | Computational methods for problems pertaining to statistics |
| 62H30 | Classification and discrimination; cluster analysis (statistical aspects) |
| 62N01 | Censored data models |
| 62P10 | Applications of statistics to biology and medical sciences; meta analysis |
Keywords:
Cox proportional hazards model; equal employment cases; frailty; Monte Carlo EM; negatively correlated processes; robustness and sensitivityReferences:
| [1] | Azzalini, A.; Capitanio, A., Distributions generated by perturbation of symmetry with emphasis on a multivariate skew-\(t\) distribution, Journal of the Royal Statistical Society: Series B, 65, 367-389, (2003) · Zbl 1065.62094 |
| [2] | Breslow, N. E., Covariance analysis of censored survival data, Biometrics, 30, 89-99, (1974) |
| [3] | Clayton, D. G., A Monte Carlo method for Bayesian inference in frailty models, Biometrics, 47, 467-485, (1991) |
| [4] | Commenges, D.; Andersen, P. K., Score test of homogeneity for survival data, Lifetime Data Analysis, 1, 145-156, (1995) · Zbl 0836.62096 |
| [5] | Duchateau, L.; Janssen, P., The frailty model, (2008), Springer New York · Zbl 1210.62153 |
| [6] | Goldstein, J., 2011. Still on Patrol After Two Decades, Valued but Rare. The New York Times, 13 December, 2011, A30. |
| [7] | Heckman, J.; Singer, B., The identifiability of the proportional hazards model, Review of Economic Studies, 51, 231-241, (1984) · Zbl 0538.90020 |
| [8] | Joly, P.; Commenges, D.; Helmer, C.; Letenneur, L., A penalized likelihood approach for an illness-death model with interval-censored data: application to age-specific incidence of dementia, Biostatistics, 3, 3, 433-443, (2002) · Zbl 1135.62394 |
| [9] | Kalbfleisch, J. D.; Prentice, R. L., The statistical analysis of failure time data, (2002), John Wiley & Sons New York · Zbl 1012.62104 |
| [10] | Kim, Y.; Kim, J.; Jang, W., An EM algorithm for the proportional hazards model with doubly censored data, Computational Statistics and Data Analysis, 57, 41-51, (2013) · Zbl 1365.62373 |
| [11] | Klein, J. P.; Moeschberger, M. L., Survival analysis techniques for censored and truncated data, (1997), Springer New York · Zbl 0871.62091 |
| [12] | Liu, L.; Huang, X., The use of Gaussian quadrature for estimation in frailty proportional hazards models, Statistics in Medicine, 27, 2665-2683, (2008) |
| [13] | Liu, L.; Wolfe, R. A.; Huang, X., Shared frailty models for recurrent events and a terminal event, Biometrics, 60, 747-756, (2004) · Zbl 1274.62827 |
| [14] | Liu, L.; Yu, Z., A likelihood reformulation method in non-normal random effects models, Statistics in Medicine, 27, 3105-3124, (2008) |
| [15] | Louis, T., Finding the observed informaiton matrix when using the EM algorithm, Journal of the Royal Statistical Society: Series B, 44, 226-233, (1982) · Zbl 0488.62018 |
| [16] | McLachlan, G.; Peel, D., Finite mixture models, (2000), Wiley New York · Zbl 0963.62061 |
| [17] | Melnykov, V.; Melnykov, I., Initializaing the EM algorithm in Gaussian mixture models with an unknown number of components, Computational Statistics and Data Analysis, 56, 1499-1510, (2012) · Zbl 1246.65025 |
| [18] | Nielsen, G. G.; Gill, R. D.; Anderson, P. K.; Sorensen, T. I.A., A counting process approach to maximum likelihood estimation in frailty models, Scandinavian Journal of Statistics, 19, 25-43, (1992) · Zbl 0747.62093 |
| [19] | O’Hagan, A.; Murphy, T. B.; Gormley, I. C., Computational aspects of Fitting mixture models via the expectation-maximization algorithm, Computational Statistics and Data Analysis, 56, 3843-3864, (2012) · Zbl 1255.62180 |
| [20] | O’Quigley, J.; Stare, J., Proportional hazards models with frailties and random effects, Statistics in Medicine, 21, 3219-3233, (2002) |
| [21] | Pan, Q.; Gastwirth, J. L., The appropriateness of survival analysis for determining lost pay in discrimination cases: application of the “lost chance” doctrine to Alexander v. milwaukee, Law, Probability & Risk, 12, 13-35, (2013) |
| [22] | Rosenbaum, P. R., Observational studies, (2002), Springer New York · Zbl 0985.62091 |
| [23] | Xu, J.; Kalbfleisch, J. D.; Tai, B., Statistical analysis of illness-death processes and semicompeting risk data, Biometrics, 66, 716-725, (2010) · Zbl 1203.62207 |
| [24] | Yue, H.; Chan, K. S., A dynamic frailty model for multivariate survival data, Biometrics, 53, 785-793, (1997) · Zbl 0890.62085 |
This reference list is based on information provided by the publisher or from digital mathematics libraries. Its items are heuristically matched to zbMATH identifiers and may contain data conversion errors. In some cases that data have been complemented/enhanced by data from zbMATH Open. This attempts to reflect the references listed in the original paper as accurately as possible without claiming completeness or a perfect matching.
