htlr_predict: Make Prediction on New Data (Advanced)

In longhaiSK/HTLR: Bayesian Logistic Regression with Heavy-Tailed Priors

Make Prediction on New Data (Advanced)

Description

This function uses MCMC samples from fitted htlrfit object OR user supplied regression coefficient to predict the class labels of test cases.

Usage

htlr_predict(
  X_ts,
  fithtlr = NULL,
  deltas = NULL,
  burn = NULL,
  thin = 1,
  usedmc = NULL,
  rep.legacy = TRUE
)

Arguments

X_ts	Matrix of values at which predictions are to be made.
fithtlr	Fitted HTLR model object.
deltas	The values of deltas (for example true deltas) used to make prediction; will override fithtlr if provided.
burn, thin	burn of Markov chain (super)iterations will be discarded for prediction, and only every thin are used.
usedmc	Indices of Markov chain iterations used for inference. If supplied, burn and thin will be ignored.
rep.legacy	To reproduce (actually incorrect) results in legacy version. See https://github.com/longhaiSK/HTLR/issues/7.

Value

A matrix of predictive probabilities, with rows for cases, cols for classes.

longhaiSK/HTLR documentation built on Nov. 15, 2024, 10:16 a.m.