# Collin Erickson

#### 7 packages on CRAN

Fits a Gaussian process model to data. Gaussian processes are commonly used in computer experiments to fit an interpolating model. The model is stored as an 'R6' object and can be easily updated with new data. There are options to run in parallel (not for Windows), and 'Rcpp' has been used to speed up calculations. Other R packages that perform similar calculations include 'laGP', 'DiceKriging', 'GPfit', and 'mlegp'.

Quickly run experiments to compare the run time and output of code blocks. The function mbc() can make fast comparisons of code, and will calculate statistics comparing the resulting outputs. It can be used to compare model fits to the same data or see which function runs faster. The function ffexp() runs a function using all possible combinations of selected inputs. This is useful for comparing the effect of different parameter values. It can also run in parallel and automatically save intermediate results, which is very useful for long computations.

Gives design points from a sequential full factorial-based Latin hypercube design, as described in Duan, Ankenman, Sanchez, and Sanchez (2015, Technometrics, <doi:10.1080/00401706.2015.1108233>).

Some R functions, such as optim(), require a function its gradient passed as separate arguments. When these are expensive to calculate it may be much faster to calculate the function (fn) and gradient (gr) together since they often share many calculations (chain rule). This package allows the user to pass in a single function that returns both the function and gradient, then splits (hence 'splitfngr') them so the results can be accessed separately. The functions provided allow this to be done with any number of functions/values, not just for functions and gradients.

Test functions are often used to test computer code. They are used in optimization to test algorithms and in metamodeling to evaluate model predictions. This package provides test functions that can be used for any purpose. Some functions are taken from <https://www.sfu.ca/~ssurjano>, but their R code is not used.

Run computer experiments using the adaptive composite grid algorithm with a Gaussian process model. The algorithm works best when running an experiment that can evaluate thousands of points from a deterministic computer simulation. This package is an implementation of a forthcoming paper by Plumlee, Erickson, Ankenman, et al. For a preprint of the paper, contact the maintainer of this package.

Provides functions for making contour plots. The contour plot can be created from grid data, a function, or a data set. If non-grid data is given, then a Gaussian process is fit to the data and used to create the contour plot.