get_sigsqs {bartMachine} | R Documentation |
Get Posterior Error Variance Estimates
Description
Returns the posterior estimates of the error variance from the Gibbs samples with an option to create a histogram of the posterior estimates of the error variance with a credible interval overlaid.
Usage
get_sigsqs(bart_machine, after_burn_in = T,
plot_hist = F, plot_CI = .95, plot_sigma = F)
Arguments
bart_machine |
An object of class “bartMachine”. |
after_burn_in |
If TRUE, only the |
plot_hist |
If TRUE, a histogram of the posterior |
plot_CI |
Confidence level for credible interval on histogram. |
plot_sigma |
If TRUE, plots |
Value
Returns a vector of posterior \sigma^2
draws (with or without the burn-in samples).
Author(s)
Adam Kapelner and Justin Bleich
See Also
Examples
## Not run:
#generate Friedman data
set.seed(11)
n = 300
p = 5
X = data.frame(matrix(runif(n * p), ncol = p))
y = 10 * sin(pi* X[ ,1] * X[,2]) +20 * (X[,3] -.5)^2 + 10 * X[ ,4] + 5 * X[,5] + rnorm(n)
##build BART regression model
bart_machine = bartMachine(X, y)
#get posterior sigma^2's after burn-in and plot
sigsqs = get_sigsqs(bart_machine, plot_hist = TRUE)
## End(Not run)
[Package bartMachine version 1.3.4.1 Index]