survival (version 2.9-6)

ridge: Ridge regression

Description

When used in a coxph or survreg model formula, specifies a ridge regression term. The likelihood is penalised by theta/2 time the sum of squared coefficients. If scale=T the penalty is calculated for coefficients based on rescaling the predictors to have unit variance. If df is specified then theta is chosen based on an approximate degrees of freedom.

Usage

ridge(..., theta, df=nvar/2, eps=0.1, scale=T)

Arguments

...

predictors to be ridged

theta

penalty is theta/2 time sum of squared coefficients

df

Approximate degrees of freedom

eps

Accuracy required for df

scale

Scale variables before applying penalty?

Value

• An object of class coxph.penalty containing the data and control functions.

References

Gray (1992) "Flexible methods of analysing survival data using splines, with applications to breast cancer prognosis" JASA 87:942--951

See Also

coxph,survreg,pspline,frailty

Examples

data(ovarian)
fit1 <- coxph(Surv(futime, fustat) ~ rx + ridge(age, ecog.ps, theta=1),
	      ovarian)
fit1
data(cancer)
lfit0 <- survreg(Surv(time, status) ~1, cancer)
lfit1 <- survreg(Surv(time, status) ~ age + ridge(ph.ecog, theta=5), cancer)
lfit2 <- survreg(Surv(time, status) ~ sex + ridge(age, ph.ecog, theta=1), cancer)
lfit3 <- survreg(Surv(time, status) ~ sex + age + ph.ecog, cancer)

lfit0
lfit1
lfit2
lfit3

Run the code above in your browser using DataCamp Workspace