lm,
glm,
censorReg,
or
censorRegList.
Calculates jackknife statistics for parameters
of interest and produces an object of class
.
jackknife(data, statistic, ...) jackknife.lm(data, statistic, ...) jackknife.glm(data, statistic, ...) jackknife.censorReg(data, statistic, ...)
censorRegList.
lm,
glm, or
censorReg objects and returns a
vector or matrix.
It may be a function (e.g.
coef) which
accepts data as the first argument;
other arguments may be passed using
args.stat.
predict(fit,newdata=orig.frame).
If the
data object is given by name (e.g.
data=fit) then use that name
in the expression,
otherwise (e.g.
data=glm(formula,dataframe)) use the name
data in
the expression, e.g.
predict(data,newdata=orig.frame).
resamp. See
for more details.
See .
# jackknife and lm
fit <- lm(Fuel~Weight+Disp., data=fuel.frame)
jackknife(fit, coef)
# jackknife and glm
fit <- glm(ozone^(1/3) ~ bs(radiation, 5,knots=185,Boundary.knots=c(7,334)) +
(wind + temperature)^2, data=air)
newdata <- air
jackknife(fit, predict(fit, newdata=newdata, type="link"))
# jackknife and censorReg (use default statistic)
fit <- censorReg(censor(days, event) ~ voltage, data = capacitor2,
weights=weights, distribution = "weibull", threshold = "Lin")
jackknife(fit, coef(fit))
# See help(bootstrap.lm) for more examples of syntax.