OptionalallowWhether a level of a grouping factor not in the fit, or a missing grouping value, is allowed (default false, an
error): that term then adds 0 for the row (the population level for that term). R's allow.new.levels.
OptionalnewdataThe rows to predict: a data frame with the columns of the fit's data that the prediction uses (the variables of the fixed-effects part and of the selected random-effects terms; not the response). Default: the rows of the fit.
OptionalnewoffsetThe offset of the rows of newdata (one number per row; default all 0) when the fit's offset was given as an
array (or not at all); an offset given to glmer as a column name is read from that column of newdata instead,
and newoffset is then a RangeError, as it is without newdata.
OptionalrandomWhether to predict the random part only, without the fixed effects and the offset (default false; R's random.only).
OptionalreWhich random-effects terms to include (R's re.form): undefined or null all of them (default); false,
"~0" or "~ 0" none (the population-level prediction); a one-sided formula such as "~ (1 | g)" those written
in it (its fixed part is ignored). Its grouping factors must be the fit's and its columns among the fit's columns
for that factor; a term may have fewer columns than the fit's.
OptionalseWhether to return the standard errors (default false; R's se.fit).
Optionaltype"link" (default) or "response": for lmer both give the same values, as in R.
Options of glmerPredict: those of LmerPredictOptions and
newoffset. Only these keys are accepted: R'sre.form,allow.new.levels,random.only,se.fit,newparams,terms,na.actionandnaActiongive the ErrorglmerPredict: argument '<key>' is not supported, any other key a RangeError.