coxme function

Fit a mixed effects Cox model

Fit a Cox model containing mixed (random and fixed) effects. Assume a Gaussian distribution for the random effects.

coxme(formula, data, weights, subset, na.action, init, control,
ties = c("efron", "breslow"),
varlist, vfixed, vinit,  x = FALSE, y = TRUE,
refine.n = 0, random, fixed, variance, ...)

Arguments

• formula: a two-sided formula with a survival object as the left hand side of a ~ operator and the fixed and random effects on the right.

• data: an optional data frame containing the variables named in the formula.

• subset, weights, na.action: further model specifications arguments as in lm; see there for details.

• init: optional initial values for the fixed effects.

• control: optional list of control options. See coxme.control for details.

• ties: method for handling exact ties in the survival time.

• varlist: the variance family to be used for each random term. If there are multiple terms it will be a list of variance functions. The default is coxmeFull. Alternatively it can be a list of matrices, in which case the coxmeMlist function is used.

• vfixed: optional named list or vector used to fix the value of one or more of the variance terms at a constant.

• vinit: optional named list or vector giving suggested starting values for the variance.

• x: if TRUE the X matrix (fixed effects) is included in the output object

• y: if TRUE the y variable (survival time) is included in the output object

• refine.n: number of samples to be used in a monte-carlo estimate of possible error in the log-likelihood of the fitted model due to inadequacy of the Laplace approximation.

• fixed, random, variance: In the preliminary version of coxme

  the fixed and random effects were separate arguments. These arguments are included for backwards compatability, but are depreciated. The variance argument is a depreciated alias for vfixed.

• ...: any other arguments are passed forward to coxme.control.

Returns

An object of class coxme.

References

S Ripatti and J Palmgren, Estimation of multivariate frailty models using penalized partial likelihood, Biometrics, 56:1016-1022, 2000.

T Therneau, P Grambsch and VS Pankratz, Penalized survival models and frailty, J Computational and Graphical Statistics, 12:156-175, 2003.

Author(s)

Terry Therneau

See Also

coxmeFull, coxmeMlist, coxme.object

Examples

# A non-significant institution effect
fit1 <- coxph(Surv(time, status) ~ ph.ecog + age, data=lung,
              subset=(!is.na(inst)))
fit2 <- coxme(Surv(time, status) ~ ph.ecog + age + (1|inst), lung)
anova(fit1, fit2)

# Shrinkage effects (equivalent to ridge regression)
temp <- with(lung, scale(cbind(age, wt.loss, meal.cal)))
rfit <- coxme(Surv(time, status) ~ ph.ecog + (temp | 1), data=lung)