Frailty.h: Fitting Frailty Models using H-likelihood Approach

In h.likelihood: Statistical Modeling and Inference via Hierarchical Likelihood

Description

Frailty.h is used to fit frailty models using h-likelihood approach. The package fits Cox's proportional hazards models with random effects (or frailties). Here, for the frailty distribution lognormal or gamma is allowed. The h-likelihood obviates the need for marginalization over the frailty distribution, leading to a statistically efficient procedure for various random-effect models including frailty models.

Usage

Frailty.h(formulaMain, censor, DataMain, RandDist = "Normal", mord = 0,
          dord = 1, Maxiter = 200, convergence = 1e-7, contrasts = NULL)

Arguments

formulaMain	Specify the formula for the mean structure of the model. e.g. y ~ x + (1|id), y: response, x: fixed covariate, id: random effect.
censor	One vector for censoring indicator (1 = uncensored, 0 = censored).
DataMain	Dataframe for formulaMain.
RandDist	Distribution for random effect ("Normal" or "Gamma").
mord	Order for the mean model (0 or 1); default = 0.
dord	Order for the dispersion components (1 or 2); default = 1.
Maxiter	Maximum number of iterations; default = 1.
convergence	Specify the convergence criterion, the default is 1e-7.
contrasts	Caution as it is currently not fully developed.

Details

Frailty.h produces estimates of fixed effects and frailty parameters as well as their standard errors. Also, Frailty.h makes it possible to fit models where the frailty distribution is not necessarily normal and estimate variance components when frailty structure is shared or nested.

Author(s)

Il Do Ha

References

Ha, I.D. and Lee, Y. (2003). Estimating frailty models via Poisson Hierarchical generalized linear models. Journal of Computational and Graphical Statistics, 12, 663-681.

Ha, I.D. and Lee, Y. (2005). Comparison of hierarchical likelihood versus orthodox best linear unbiased predictor approaches for frailty models. Biometrika, 92, 717-723.

Ha, I.D., Lee, Y. and Song, J.-K. (2001). Hierarchical likelihood approach for frailty models. Biometrika, 88, 233-243.

Lee, Y., Nelder, J.A., and Pawitan, Y. (2006). Generalized Linear Models with Random Effects. Boca Raton: Chapman & Hall/CRC.

Examples

## Not run: 
data(cgd)
data(kidney)

#### Frailty model

#### Analysis of Kidney data
kln11 <- Frailty.h(time ~ sex + age + (1|patient)-1, kidney$censor, kidney,
                   RandDist = "Normal", mord = 1, dord = 1)
kln12 <- Frailty.h(time ~ sex + age + (1|patient) - 1, kidney$censor, kidney,
                   RandDist = "Normal", mord = 1, dord = 2)
kg11 <- Frailty.h(time ~ sex + age + (1|patient) - 1, kidney$censor, kidney, 
                  RandDist = "Gamma", mord = 1, dord = 1)
kg12 <- Frailty.h(time ~ sex + age + (1|patient) - 1, kidney$censor, kidney,
                  RandDist = "Gamma", mord = 1, dord = 2)

#### Analysis of CGD data
cgd11 <- Frailty.h(TIME ~ TRT + (1|HOSPITAL) + (1|PATIENT) - 1, cgd$DEL, cgd,
                   RandDist = "Normal", mord = 1, dord = 1)
cgd12 <- Frailty.h(TIME ~ TRT + (1|HOSPITAL) + (1|PATIENT) - 1, cgd$DEL, cgd,
                   RandDist = "Normal", mord = 1, dord = 2)


## End(Not run)           

h.likelihood documentation built on May 2, 2019, 4:36 p.m.