predict.mer: A tentative version of predict for mer objects

In gmonette/spida15: Collection of miscellaneous functions for mixed models etc. prepared for SPIDA 2009+ (development version)

Usage

predict.mer(model, data = model.matrix(model), form, verbose = FALSE)

Arguments

model
data
form
verbose

Examples

##---- Should be DIRECTLY executable !! ----
##-- ==>  Define data, use random,
##--	or do  help(data=index)  for the standard data sets.

## The function is currently defined as
function (model, data = model.matrix(model), form, verbose = FALSE) 
{
    help = "\n    This is a crude predict method for class 'mer'\n    Limitations:\n    1) When invoked on the model it only returns the linear predictor for the\n       complete data. Note that a complete data frame is easily obtained with\n       model.frame( model )\n    2) When the 'data' argument is provided for 'new' data, it is also\n       necessary to provide the correct rhs of the fixed effects formula.\n\n    "
    if (missing(form) & !missing(data)) {
        cat(help)
        stop("Need 'form' if 'data' given")
    }
    if (!missing(data)) {
        data = model.matrix(form, data)
        cnames = colnames(data)
        if (verbose) 
            print(cnames)
        fnames = names(fixef(model))
        if (verbose) 
            print(fnames)
        if (any(cnames != fnames)) {
            cat("\nMatrix names:\n")
            print(cnames)
            cat("\nCoeff names:\n")
            print(fnames)
            warning("matrix and coeff names not the same")
        }
    }
    data %*% fixef(model)
  }

gmonette/spida15 documentation built on May 17, 2019, 7:26 a.m.