predict.pibblefit: Predict response from new data

In jsilve24/mongrel: Bayesian Multinomial Logistic Normal Regression

Description

Predict response from new data

Usage

## S3 method for class 'pibblefit'
predict(object, newdata = NULL,
  response = "LambdaX", size = NULL, use_names = TRUE,
  summary = FALSE, iter = NULL, from_scratch = FALSE, ...)

Arguments

object	An object of class pibblefit
newdata	An optional matrix for which to evaluate predictions. If NULL (default), the original data of the model is used.
response	Options = "LambdaX":Mean of regression, "Eta", "Y": counts
size	the number of counts per sample if response="Y" (as vector or matrix), default if newdata=NULL and response="Y" is to use colsums of m$Y. Otherwise uses median colsums of m$Y as default. If passed as a matrix should have dimensions ncol(newdata) x iter.
use_names	if TRUE apply names to output
summary	if TRUE, posterior summary of predictions are returned rather than samples
iter	number of iterations to return if NULL uses object$iter
from_scratch	should predictions of Y come from fitted Eta or from predictions of Eta from posterior of Lambda? (default: false)
...	other arguments passed to summarise_posterior

Details

currently only implemented for pibblefit objects in coord_system "default" "alr", or "ilr".

Value

(if summary==FALSE) array D x N x iter; (if summary==TRUE) tibble with calculated posterior summaries

Examples

sim <- pibble_sim()
fit <- pibble(sim$Y, sim$X)
predict(fit)[,,1:2] # just show 2 samples

jsilve24/mongrel documentation built on Jan. 27, 2022, 9:54 p.m.