ks_test {longevity} | R Documentation |
Goodness-of-fit diagnostics
Description
Warning: EXPERIMENTAL Compute the Kolmogorov-Smirnov test statistic and compare it with a simulated null distribution obtained via a parametric bootstrap.
Usage
ks_test(
time,
time2 = NULL,
event = NULL,
thresh = 0,
ltrunc = NULL,
rtrunc = NULL,
type = c("right", "left", "interval", "interval2"),
family = c("exp", "gp", "gomp", "gompmake", "weibull", "extgp", "gppiece",
"extweibull", "perks", "beard", "perksmake", "beardmake"),
B = 999L,
arguments = NULL,
...
)
Arguments
time |
excess time of the event of follow-up time, depending on the value of event |
time2 |
ending excess time of the interval for interval censored data only. |
event |
status indicator, normally 0=alive, 1=dead. Other choices are |
thresh |
vector of thresholds |
ltrunc |
lower truncation limit, default to |
rtrunc |
upper truncation limit, default to |
type |
character string specifying the type of censoring. Possible values are " |
family |
string; choice of parametric family |
B |
number of bootstrap simulations |
arguments |
a named list specifying default arguments of the function that are common to all |
... |
additional parameters, currently ignored |
Value
a list with elements
-
stat
the value of the test statistic -
pval
p-value obtained via simulation
Note
The bootstrap scheme requires simulating new data, fitting a parametric model and estimating the nonparametric maximum likelihood estimate for each new sample. This is computationally intensive in large samples.