Predicted values based on a model object produced by selm with
univariate response.
# S3 method for selm
predict(object, newdata, param.type = "CP",
interval = "none", level = 0.95, na.action = na.pass, ...)a vector of predictions (if interval="none") or a matrix of
predictions and bounds with column names fit, lwr, and
upr, if interval is set.
an object of class selm as produced by a call to
function selm with univariate response.
an optional data frame in which to look for variables with which to predict. If omitted, the fitted values are used.
a character string with the required parameterization; it must be one of
"CP", "DP", "pseudo-CP", or possibly their
equivalent lowercase.
type of interval calculation among "none", "confidence",
"prediction"; it can be abbreviated.
tolerance/confidence level (default value is 0.95).
function determining what should be done with missing
values in newdata. The default is to predict NA.
not used, only there for compatibility reasons.
Adelchi Azzalini
Predicted values are obtained by evaluating the regression function in the
dataframe newdata (which defaults to model.frame(object)).
Setting interval other than "none" produces computation of
confidence or prediction (tolerance) intervals at the specified level.
If newdata is omitted the predictions are based on the data used for
the fit.
The action taken in case of missing data is regulated by argument
na.action, along the lines of function predict.lm.
A detailed description of the methodology underlying predict.selm
is available in the technical note of Azzalini (2016).
Azzalini, A. (2016).
Derivation of various types of intervals from a selm object.
Technical note distributed with the documentation of the R package
sn, in file
selm-intervals.pdf within section
‘User guide, package vignettes and other documentation’.
data(barolo)
attach(barolo)
A75 <- (reseller=="A" & volume==75)
detach(barolo)
m3 <- selm(log(price, 10) ~ age, data=barolo[A75,], family="ST")
Run the code above in your browser using DataLab