| predict.zlm {BMS} | R Documentation |
Expected value (And standard errors) of predictions based on 'zlm' linear Bayesian model under Zellner's g prior
## S3 method for class 'zlm': ## S3 method for class 'zlm': predict(object, newdata = NULL, se.fit = FALSE, ...)
object |
a zlm linear model object - see zlm |
newdata |
An optional data.frame, matrix or vector containing variables with which to predict. If omitted, then (the expected values of) the fitted values are returned. |
se.fit |
A switch indicating if the standard deviations for the predicted varaibles are required. |
... |
further arguments passed to or from other methods. |
A vector with (expected values of) fitted values.
If se.fit is TRUE, then the output is a list with the following elements:
fit |
a vector with the expected values of fitted values |
std.err |
a vector with the standard deviations of fitted values |
se.fit |
a vector with the standard errors without the residual scale akin to se.fit in predict.lm |
residual.scale |
The part from the standard deviations that involves the identity matrix. Note that sqrt(se.fit^2+residual.scale^2) yields std.err. |
Martin Feldkircher and Stefan Zeugner
bms for creating zlm objects, predict.lm for a comparable function, predict.bma for predicting with bma objects
Check http://bms.zeugner.eu for additional help.
data(datafls) mm=zlm(datafls,g="EBL") predict(mm) #fitted values predict(mm, newdata=1:41) #prediction based on a 'new data point' #prediction based on a 'new data point', with 'standard errors' predict(mm, newdata=datafls[1,], se.fit=TRUE)