nlme (version 3.1-86)

predict.lme: Predictions from an lme Object

Description

The predictions at level $i$ are obtained by adding together the population predictions (based only on the fixed effects estimates) and the estimated contributions of the random effects to the predictions at grouping levels less or equal to $i$. The resulting values estimate the best linear unbiased predictions (BLUPs) at level $i$. If group values not included in the original grouping factors are present in newdata, the corresponding predictions will be set to NA for levels greater or equal to the level at which the unknown groups occur.

Usage

## S3 method for class 'lme':
predict(object, newdata, level, asList, na.action, \dots)

Value

if a single level of grouping is specified in level, the returned value is either a list with the predictions split by groups (asList = TRUE) or a vector with the predictions (asList = FALSE); else, when multiple grouping levels are specified in level, the returned object is a data frame with columns given by the predictions at different levels and the grouping factors.

See Also

lme, fitted.lme

Examples

fm1 <- lme(distance ~ age, Orthodont, random = ~ age | Subject)
newOrth <- data.frame(Sex = c("Male","Male","Female","Female","Male","Male"),
                      age = c(15, 20, 10, 12, 2, 4),
                      Subject = c("M01","M01","F30","F30","M04","M04"))
predict(fm1, newOrth, level = 0:1)