Larson and Dinse (1985) have introduced the mixture model as an additional competing risks model. In the same article, the authors have suggested that this model can be upscaled to handle the presence of missing failure causes in data. We respond to this proposal in this article and develop a regression model for analysis of data that comes with this complication. We also demonstrate that, with minimal adjustments, the proposed model can be applied in discrete time. This development will be of benefit to discrete time competing risks as analysis of data with this complication is a subject that has not received adequate attention. The mixture model has two components, the incidence and the latency component. It is demonstrated that the parameters related to the model for the latency component as proposed by Larson and Dinse (1985) can be estimated by applying a certain Poisson regression.
AmbrogiF.BiganzoliE., & BoracchiP. (2009). Estimating Crude Cumulative incidences through Multinomial Logit Regression on Discrete Cause Specific Hazard.Computational Statistics and Data Analysis, 53, 2767-2779.
2.
AndersonJ.GoetghebeurE., & RyanL. (1996). Missing cause of death information in the analysis of survial data.Statistics in Medicine, 16, 2191-2201.
3.
BeggC.B., & GrayR. (1984). Calculation of polytomous logistic regression parameters using individualized regressions.Biometrika, 71, 11-18.
4.
BetancurM.M. (2013). Regression modeling with missing outcomes: competing risks and longitudinal data.PhD thesis, Universite Paris Sud.
5.
ChoiC. (2002). On the Mixture Models in Survival Analysis with Competing Risks and Covariates. PhD thesis, Hong Kong polytechnic University.
6.
CoxD. (1972). Regression Models and Life Tables.Journal of Royal Statistical Society B, 34, 187-220.
7.
CroissantY., & GravesS. (2020). Ecdat: Data Sets for Econometrics. R package version 0.3-7.
8.
DewanjiA. (1992). A note on a test for competing risks with missing failure.Biometrica, 79, 855-857.
9.
DinseG.E. (1982). Nonparametric estimation for partially-complete time and of failure data.Biometrics, 38, 417-431.
10.
EscaralaG., & BowaterR.J. (2008). FittiNG A SEMI-PARAMETRIC MIXTURE MODEL FOR COMPETING RISKS IN SURVIVAL Data.Communications in Statistics – Theory and Methods, 37, 277-293.
11.
GoetghebeurE., & RyanL. (1990). A modified logrank test for competing risks with missing failure type.Biometrika, 77, 207-211.
12.
GoetghebeurE., & RyanL. (1995). Analysis of competing risks survival data when some failure types are missing.Biometrika, 82, 821-834.
13.
HallerB. (2014). The Analysis of Competing Risks Data with a Focus on Estimation of Cause-Specific and Subdistribution Hazard Ratios from a Mixture Model. PhD thesis, Ludwig-Maximiliäns-Universitat München.
14.
HolfordT. (1980). The Analysis of rates and of survivorship using log-linear models.Biometrics, 36, 299-305.
15.
LairdN., & OliverD. (1981). Covariance analysis of censored survival data using log-linear analysis techniques.Biometrica, 76, 231-240.
16.
LarsonM.G., & DinseG.E. (1985). A mixture model for the regression analysis of competing risks data.Journal of the Royal Statistical Society. Series C (Applied Statistics), 34, 201-211.
17.
LauB.ColeS., & GangeS. (2011). Parametric mixture models to evaluate and summarize hazard ratios in the presence of competing risks with time-dependent hazards and delayed entry.Statistic in Medicine, 30, 654-665.
18.
LauB.ColeS.R.MooreR.D., & GangeS.J. (2008). Evaluating competing adverse and beneficial outcomes using a mixture model.Stat Med, 27, 4313-4313.
19.
LeeM.FeuerE., & FineJ. (2018). On the analysis of discrete time competing risks data. Biometrics. doi: 10.1111/biom.12881.
20.
LuK., & TsiatisA. (2001). Multiple imputation methods for estimating regression coefficients in the competing risks model with missing cause of failure.Biometrics, 57, 1191-1197.
21.
MallerR., & ZhouX. (2002). Analysis of parametric models for competing risks.Statistica Sinica, 12, 725-750.
NgS.McLachlanG.YauK.W., & LeeA. (2004). Modelling the distribution of ischaemic stroke-specific survival time using an em-based mixture approach with random effects adjustment.Stat Med, 23, 2729-2744.
24.
NgS.K., & McLachanG.J. (2003). An EM-based semi-parametric mixture model approach to the regression analysis of competing-risks data.Stat Med, 22, 1097-1111.
25.
NicolaieM.van HouwelingenH.C., & PutterH. (2010). Vertical modeling: A pattern mixture approach for competing risks modeling.Statistics in Medicine, 29, 1190-1205.
26.
NicolaieM.A.TaylorJ., & LegrandC. (2018). Vertical modeling: analysis of competing risks data with a cure fraction. Lifetime Data Analysis.
27.
NicolaieM.A.van HouwelingenH.C., & PutterH. (2015). Vertical modelling: Analysis of competing risks data with missing causes of failure.Statistical Methods in Medicine, 29, 1190-1205.
28.
PengY., & TaylorJ. (2014). Cure Models, chapter 6, pages 113–134. Chapman and Hall, Boca Raton, USA.
29.
PrenticeR.L.KalbfleischJ.D.PetersonA.V.FlournoyN.FarewellV.T., & BreslowN.E. (1978). The analysis of failure times in the presence of competing risks.Biometrics, 34, 541-554.
30.
RubinD.B. (1976). Inference and missing data.Biometrika, 63, 581-592.
31.
TutzG. (1995). Competing risks models in discrete time with nominal or ordinal categories of response.Quality and Quantity, 29, 405-420.
ZhipingT. (2011). On Competing Risks data with covariates and Long-term Survivors. PhD thesis, Hong Kong Polytechnic University.
34.
NdlovuB.MelesseS., & ZewotirT. (2019). A mixture model with application to discrete competing risks data. South African Journal of Statistics, 2, 73-86.
