Engineering topics
Rumsey, Kelin
Publications and source records attributed to Rumsey, Kelin.
Concordance
Dimension reduction techniques have long been an important topic in statistics, and active subspaces (AS) have received much attention this past decade in the computer experiments literature. The most common approach towards estimating the AS is to use Monte Carlo with numerical gradient evaluation. This software takes a new approach towards estimating the active subspace, in which Multivariate Adaptive Regression Splines (MARS) is used as a surrogate model, and relevant calculation are obtained analytically. Using a MARS surrogate has many advantages including improved scaling, better estimation of active subspaces in high dimensions and the ability to handle a large number of prior distributions in closed form. This software is also capable of conducting a so-called "concordance analysis", using a generalization of ASs which we refer to as "Co-Active" subspaces.
BPP R package
The BPP package contains R code to perform Bayesian projection pursuit regression under various regression settings. Bayesian projection pursuit regression learns the functional relationship between inputs and outputs by considering a sum of functions of one dimensional projections of the inputs. This is useful as a general machine learning software.
Embracing unidentifiability in Bayesian model calibration with modularization.
Abstract not provided.
Robust Inference for Physical Parameters with Bayesian Model Calibration.
Abstract not provided.
"Learn something new: techniques for broadening your statistical skillset".
Abstract not provided.
Classification and Outlier Detection with Functional Extremal Depth.
Abstract not provided.
Temporal Modeling of Heavy-Tailed Data via Non-Gaussian Latent Process.
Abstract not provided.