predict.face.sparse: Subject-specific curve prediction from a face.sparse fit

In lxiao5/face: Fast Covariance Estimation for Sparse Functional Data

Subject-specific curve prediction from a face.sparse fit

Description

Predict subject-specific curves based on a fit from "face.sparse".

Usage

## S3 method for class 'face.sparse'
predict(object, newdata,...)

Arguments

object	a fitted object from the R function "face.sparse".
newdata	a data frame with three arguments: (1) argvals: observation times; (2) subj: subject indices; (3) y: values of observations. NA values are allowed in "y" but not in the other two.
...	further arguments passed to or from other methods.

Details

This function makes prediction based on observed data for each subject. So for each subject, it requires at least one observed data. For the time points prediction is desired but no observation is available, just make the corresponding data$y as NA.

Value

object	A "face.sparse" fit
newdata	Input
y.pred,mu.pred,Chat.pred, Chat.diag.pred, var.error.pred	Predicted/estimated objects at the observation time points in newdata
rand_eff	if calculate.scores in object is TRUE (typically FALSE), then predicted scores rand_eff$scores will be calculated.
...	...

Author(s)

Luo Xiao <[e-mail]>

References

Luo Xiao, Cai Li, William Checkley and Ciprian Crainiceanu, Fast covariance estimation for sparse functional data, Stat. Comput., doi: 10.1007/s11222-017-9744-8.

Examples

#See the examples for "face.sparse".

lxiao5/face documentation built on July 23, 2022, 6:24 p.m.