An EM algorithm for the estimation of parameters of a flexible cure rate model with generalized gamma lifetime and model discrimination using likelihood- and information-based methods

In this paper, we consider the Conway–Maxwell Poisson (COM-Poisson) cure rate model based on a competing risks scenario. This model includes, as special cases, some of the well-known cure rate models discussed in the literature. By assuming the time-to-event to follow the generalized gamma distribution, which contains some of the commonly used lifetime distributions as special cases, we develop exact likelihood inference based on the expectation maximization algorithm. The standard errors of the maximum likelihood estimates are obtained by inverting the observed information matrix. An extensive Monte Carlo simulation study is performed to examine the method of inference developed here. Model discrimination within the generalized gamma family is also carried out by means of likelihood- and information-based methods to select the particular lifetime distribution that provides an adequate fit to the data. Finally, a data on cancer recurrence is analyzed to illustrate the flexibility of the COM-Poisson family and the generalized gamma family so as to select a parsimonious competing cause distribution and a lifetime distribution that jointly provide an adequate fit to the data.

[1]  J W Denham,et al.  A generalized F mixture model for cure rate estimation. , 1998, Statistics in medicine.

[2]  Josemar Rodrigues,et al.  A Bayesian destructive weighted Poisson cure rate model and an application to a cutaneous melanoma data , 2012, Statistical methods in medical research.

[3]  N. Balakrishnan,et al.  Generalized gamma frailty model , 2006, Statistics in medicine.

[4]  Paul Janssen,et al.  One-sided tests in shared frailty models , 2008 .

[5]  N. L. Johnson,et al.  Continuous Univariate Distributions. , 1995 .

[6]  Joseph G. Ibrahim,et al.  A New Bayesian Model For Survival Data With a Surviving Fraction , 1999 .

[7]  Khodabina Morteza,et al.  SOME PROPERTIES OF GENERALIZED GAMMA DISTRIBUTION , 2010 .

[8]  Narayanaswamy Balakrishnan,et al.  COM–Poisson cure rate survival models and an application to a cutaneous melanoma data , 2009 .

[9]  Shyam L. Kalla,et al.  A generalized gamma distribution and its application in reliabilty , 1996 .

[10]  Galit Shmueli,et al.  Conjugate Analysis of the Conway-Maxwell-Poisson Distribution , 2006 .

[11]  G. McLachlan,et al.  The EM algorithm and extensions , 1996 .

[12]  Narayanaswamy Balakrishnan,et al.  Correlated destructive generalized power series cure rate models and associated inference with an application to a cutaneous melanoma data , 2012, Comput. Stat. Data Anal..

[13]  Joseph G. Ibrahim,et al.  Cure rate models: A unified approach , 2005 .

[14]  Bradley P Carlin,et al.  Flexible Cure Rate Modeling Under Latent Activation Schemes , 2007, Journal of the American Statistical Association.

[15]  Joseph G. Ibrahim,et al.  Bayesian Survival Analysis , 2004 .

[16]  Narayanaswamy Balakrishnan,et al.  Destructive weighted Poisson cure rate models , 2011, Lifetime data analysis.

[17]  T. Minka,et al.  A useful distribution for fitting discrete data: revival of the Conway–Maxwell–Poisson distribution , 2005 .

[18]  G. McLachlan,et al.  The EM Algorithm and Extensions: Second Edition , 2008 .

[19]  J. Lawless.  Statistical Models and Methods for Lifetime Data , 2002 .

[20]  J. Boag,et al.  Maximum Likelihood Estimates of the Proportion of Patients Cured by Cancer Therapy , 1949 .

[21]  Narayanaswamy Balakrishnan,et al.  Expectation maximization-based likelihood inference for flexible cure rate models with Weibull lifetimes , 2016, Statistical methods in medical research.

[22]  V. Farewell,et al.  The use of mixture models for the analysis of survival data with long-term survivors. , 1982, Biometrics.

[23]  K. Yamaguchi.  Accelerated Failure-Time Regression Models with a Regression Model of Surviving Fraction: An Application to the Analysis of “Permanent Employment” in Japan , 1992 .

[24]  K. Liang,et al.  Asymptotic Properties of Maximum Likelihood Estimators and Likelihood Ratio Tests under Nonstandard Conditions , 1987 .

[25]  A Yu Yakovlev,et al.  Stochastic Models of Tumor Latency and Their Biostatistical Applications , 1996 .

[26]  R. Prentice.  A LOG GAMMA MODEL AND ITS MAXIMUM LIKELIHOOD ESTIMATION , 1974 .

[27]  K. Lange.  A gradient algorithm locally equivalent to the EM algorithm , 1995 .

[28]  N. Balakrishnan,et al.  Lognormal lifetimes and likelihood-based inference for flexible cure rate models based on COM-Poisson family , 2013, Comput. Stat. Data Anal..

[29]  J. Kalbfleisch,et al.  The Statistical Analysis of Failure Time Data , 1980 .

[30]  J. P. Sy,et al.  Estimation in a Cox Proportional Hazards Cure Model , 2000, Biometrics.

[31]  D. Cox,et al.  Analysis of Survival Data. , 1985 .

[32]  T. Louis.  Finding the Observed Information Matrix When Using the EM Algorithm , 1982 .

[33]  W. Nelson.  Statistical Methods for Reliability Data , 1998 .

[34]  V. Sondak,et al.  High- and low-dose interferon alfa-2b in high-risk melanoma: first analysis of intergroup trial E1690/S9111/C9190. , 2000, Journal of clinical oncology : official journal of the American Society of Clinical Oncology.

[35]  Anthony Y. C. Kuk,et al.  A mixture model combining logistic regression with proportional hazards regression , 1992 .

[36]  J. Kalbfleisch,et al.  The Statistical Analysis of Failure Time Data: Kalbfleisch/The Statistical , 2002 .

[37]  J. Kalbfleisch,et al.  The Statistical Analysis of Failure Time Data , 1980 .

[38]  E. Stacy.  A Generalization of the Gamma Distribution , 1962 .

[39]  Jim E. Griffin,et al.  A Bayesian Partition Model for Customer Attrition , 2001 .

[40]  Joseph Berkson,et al.  Survival Curve for Cancer Patients Following Treatment , 1952 .

[41]  D. Rubin,et al.  Maximum likelihood from incomplete data via the EM - algorithm plus discussions on the paper , 1977 .