calc_pred_moments: Calculate Predictive Moments

In shrinkGPR: Scalable Gaussian Process Regression with Hierarchical Shrinkage Priors

Calculate Predictive Moments

Description

calc_pred_moments calculates the predictive means and variances for a fitted shrinkGPR model at new data points.

Usage

calc_pred_moments(object, newdata, nsamp = 100)

Arguments

object	A shrinkGPR object representing the fitted Gaussian process regression model.
newdata	Optional data frame containing the covariates for the new data points. If missing, the training data is used.
nsamp	Positive integer specifying the number of posterior samples to use for the calculation. Default is 100.

Details

This function computes predictive moments by marginalizing over posterior samples from the fitted model. If the mean equation is included in the model, the corresponding covariates are used.

Value

A list with two elements:

• means: A matrix of predictive means for each new data point, with the rows being the samples and the columns the data points.

• vars: An array of covariance matrices, with the first dimension corresponding to the samples and second and third dimensions to the data points.

Examples

if (torch::torch_is_installed()) {
  # Simulate data
  set.seed(123)
  torch::torch_manual_seed(123)
  n <- 100
  x <- matrix(runif(n * 2), n, 2)
  y <- sin(2 * pi * x[, 1]) + rnorm(n, sd = 0.1)
  data <- data.frame(y = y, x1 = x[, 1], x2 = x[, 2])

  # Fit GPR model
  res <- shrinkGPR(y ~ x1 + x2, data = data)

  # Calculate predictive moments
  momes <- calc_pred_moments(res, nsamp = 100)
  }

shrinkGPR documentation built on April 4, 2025, 3:07 a.m.