Compute deviance for Cox model
Compute the deviance (-2 log partial likelihood) for Cox model.
coxnet.deviance( pred = NULL, y, x = NULL, offset = NULL, weights = NULL, std.weights = TRUE, beta = NULL )
pred: Fit vector or matrix (usually from glmnet at a particular lambda or a sequence of lambdas).y: Survival response variable, must be a Surv or stratifySurv object.x: Optional x matrix, to be supplied if pred = NULL.offset: Optional offset vector.weights: Observation weights (default is all equal to 1).std.weights: If TRUE (default), observation weights are standardized to sum to 1.beta: Optional coefficient vector/matrix, to be supplied if pred = NULL.A vector of deviances, one for each column of predictions.
Computes the deviance for a single set of predictions, or for a matrix of predictions. The user can either supply the predictions directly through the pred option, or by supplying the x matrix and beta coefficients. Uses the Breslow approach to ties.
The function first checks if pred is passed: if so, it is used as the predictions. If pred is not passed but x and beta
are passed, then these values are used to compute the predictions. If neither x nor beta are passed, then the predictions are all taken to be 0.
coxnet.deviance() is a wrapper: it calls the appropriate internal routine based on whether the response is right-censored data or (start, stop] survival data.
set.seed(1) eta <- rnorm(10) time <- runif(10, min = 1, max = 10) d <- ifelse(rnorm(10) > 0, 1, 0) y <- survival::Surv(time, d) coxnet.deviance(pred = eta, y = y) # if pred not provided, it is set to zero vector coxnet.deviance(y = y) # example with x and beta x <- matrix(rnorm(10 * 3), nrow = 10) beta <- matrix(1:3, ncol = 1) coxnet.deviance(y = y, x = x, beta = beta) # example with (start, stop] data y2 <- survival::Surv(time, time + runif(10), d) coxnet.deviance(pred = eta, y = y2) # example with strata y2 <- stratifySurv(y, rep(1:2, length.out = 10)) coxnet.deviance(pred = eta, y = y2)
coxgrad