# GPM v3.0.0

0

0th

Percentile

## Gaussian Process Modeling of Multi-Response and Possibly Noisy Datasets

Provides a general and efficient tool for fitting a response surface to a dataset via Gaussian processes. The dataset can have multiple responses and be noisy (with stationary variance). The fitted GP model can predict the gradient as well. The package is based on the work of Bostanabad, R., Kearney, T., Tao, S. Y., Apley, D. W. & Chen, W. (2018) Leveraging the nugget parameter for efficient Gaussian process modeling. International Journal for Numerical Methods in Engineering, 114, 501-516.

## Functions in GPM

 Name Description Fit The Fitting Function of GPM Package Predict The Prediction Function of GPM Package Auxil An auxiliary function used in calculating the negative log-likelehood and its gradient MatrixAlgebra A Set of Functions for Doing Some Calculations on Matrices in GPM Package NLogL_G The Function for calculating the gradient of Negative Log-Likelehood in GPM Package NLogL The Function for calculating the Negative Log-Likelehood in GPM Package Draw The Plotting Function of GPM Package CorrMat Two Functions for Constructing the Correlation Matrix in GPM Package No Results!