calibration {MachineShop} | R Documentation |
Calculate calibration estimates from observed and predicted responses.
calibration( x, y = NULL, breaks = 10, span = 0.75, dist = NULL, na.rm = TRUE, ... )
x |
observed responses or resample result containing observed and predicted responses. |
y |
predicted responses if not contained in |
breaks |
value defining the response variable bins within which to
calculate observed mean values. May be specified as a number of bins, a
vector of breakpoints, or |
span |
numeric parameter controlling the degree of loess smoothing. |
dist |
character string specifying a distribution with which to estimate
observed survival means. Possible values are |
na.rm |
logical indicating whether to remove observed or predicted
responses that are |
... |
arguments passed to other methods. |
Calibration
class object that inherits from data.frame
.
## Requires prior installation of suggested package gbm to run library(survival) res <- resample(Surv(time, status) ~ ., data = veteran, model = GBMModel, control = CVControl(times = c(90, 180, 360))) cal <- calibration(res) plot(cal)