predict: Posterior predictive samples from models fit by 'bayesGAM',...

In bayesGAM: Fit Multivariate Response Generalized Additive Models using Hamiltonian Monte Carlo

Posterior predictive samples from models fit by bayesGAM, but with new data

Description

Draw from the posterior predictive distribution applied to new data

Usage

## S4 method for signature 'bayesGAMfit'
predict(object, newdata, draws = NULL, ...)

Arguments

object	Object of type bayesGAMfit generated from bayesGAM.
newdata	A data frame with new data applied to the bayesGAMfit object
draws	An integer indicating the number of draws to return. The default and maximum number of draws is the size of the posterior sample.
...	Additional arguments for postrior_predict

Value

a list of D by N matrices, where D is the number of draws from the posterior predictive distribution and N is the number of data points being predicted per draw.

References

Goodrich B, Gabry J, Ali I & Brilleman S. (2020). rstanarm: Bayesian applied regression modeling via Stan. R package version 2.19.3 https://mc-stan.org/rstanarm.

Jonah Gabry, Ben Goodrich and Martin Lysy (2020). rstantools: Tools for Developing R Packages Interfacing with 'Stan'. https://mc-stan.org/rstantools/, https://discourse.mc-stan.org/.

Examples

set.seed(432)
f <- bayesGAM(weight ~ np(height), data = women,
              family = gaussian, iter=500, chains = 1)
newheights <- with(women, rnorm(10, mean = mean(height)), sd=sd(height))
women2 <- data.frame(height=newheights)

pred <- predict(f, women2, draws=100)

bayesGAM documentation built on March 18, 2022, 6:29 p.m.