| coxlps.object {blapsr} | R Documentation |
Object from a Cox proportional hazards fit with Laplace-P-splines.
Description
An object returned by the coxlps function consists in a list
with various components related to the fit of a Cox model using the
Laplace-P-spline methodology.
Value
A coxlps object has the following elements:
formula |
The formula of the Cox model. |
K |
Number of B-spline basis functions used for the fit. |
penalty.order |
Chosen penalty order. |
latfield.dim |
The dimension of the latent field. This is equal to the sum of the number of B-spline coefficients and the number of regression parameters related to the covariates. |
n |
Sample size. |
num.events |
The number of events that occurred. |
event.times |
The standardized event times, i.e. if t denotes
the original time scale, then |
tup |
The upper bound of the follow-up, i.e. |
sd.time |
The standard deviation of the event times in original scale. |
event.indicators |
The event indicators. |
regcoeff |
Posterior estimates of the regression coefficients. coef gives the point estimate, sd.post gives the posterior standard deviation, z is the Wald test statistic, lower .95 and upper .95 the posterior approximate 95% quantile-based credible interval. |
penalty.vector |
The selected grid of penalty values. |
vmap |
The maximum a posteriori of the (log) penalty parameter. |
spline.estim |
The estimated B-spline coefficients. |
edf |
Estimated effective degrees of freedom for each latent field variable. |
ED |
The effective model dimension. |
Covthetamix |
The posterior covariance matrix of the B-spline coefficients. |
X |
The matrix of covariate values. |
loglik |
The log-likelihood evaluated at the posterior latent field estimate. |
p |
Number of parametric coefficients in the model. |
AIC.p |
The AIC computed with the formula -2*loglik+2*p, where p is the number of parametric coefficients. |
AIC.ED |
The AIC computed with the formula -2*loglik+2*ED, where ED is the effective model dimension. |
BIC.p |
The BIC computed with the formula -2*loglik+p*log(ne), where p is the number of parametric coefficients and ne the number of events. |
BIC.ED |
The BIC computed with the formula -2*loglik+ED*log(ne), where ED is the effective model dimension and ne the number of events. |
Author(s)
Oswaldo Gressani oswaldo_gressani@hotmail.fr.