predict.psFit {fitPS} | R Documentation |
S3 predict method for an object of class psFit
Description
S3 predict method for an object of class psFit
Usage
## S3 method for class 'psFit'
predict(
object,
newdata,
interval = c("none", "prof", "wald"),
level = 0.95,
...
)
Arguments
object |
an object of class psFit , usually from fitDist
or fitZIDist .
|
newdata |
an optional vector of integers at which to calculate
\Pr(X = x) .
|
interval |
either "none" , "prof" , or "wald" and can
be abbreviated. If "prof" or "wald" AND the zeta model has
been used then an interval, based on the bounds of a 100 * level
confidence interval for the shape parameter, is given for each predicted
probability. The interval is provided based on either a Profile Likelihood,
or a Wald, confidence interval for the shape, and therefore cannot really
be regarded as a confidence interval for the probabilities. The intervals
might be more sensibly regarded as a measure of how sensitive the
probabilities are to the choice of shape parameter. NOTE: this parameter is
ignored if the Zero-inflated (ZIZ) model has been used.
|
level |
the level of a confidence interval. Ignored if interval ==
"none" .
|
... |
other arguments passed to predict —not used
|
Value
either a named vector of fitted probabilities, or a data.frame
with columns predicted
, lower
, and upper
and the row
names set to show what terms are being calculatd
Examples
data(Psurveys)
roux = Psurveys$roux
fit = fitDist(roux)
predict(fit, interval = "prof")
[Package
fitPS version 1.0.1
Index]