5 papers
Multivariate Varying-Coefficient BART with Graphical Horseshoe Priors
Soham Ghosh, Sameer K. Deshpande
Modern multivariate regression problems involve several related outcomes whose regression effects are not only nonlinear, heterogeneous, and outcome-specific, but also where the re…
High-dimensional regression with outcomes of mixed-type using the multivariate spike-and-slab LASSO
Soham Ghosh, Sameer K. Deshpande
We consider a high-dimensional multi-outcome regression in which possibly dependent, binary and continuous outcomes are regressed onto covariates. We model the observed ou…
Fitting sparse high-dimensional varying-coefficient models with Bayesian regression tree ensembles
Soham Ghosh, Saloni Bhogale, Sameer K. Deshpande
By allowing the effects of covariates in a linear regression model to vary as functions of additional effect modifiers, varying-coefficient models (VCMs) strike a compellin…
Scalable piecewise smoothing with BART
Ryan Yee, Soham Ghosh, Sameer K. Deshpande
Although it is an extremely effective, easy-to-use, and increasingly popular tool for nonparametric regression, the Bayesian Additive Regression Trees (BART) model is limited by th…
Spatial Dependencies in Item Response Theory: Gaussian Process Priors for Geographic and Cognitive Measurement
Mingya Huang, Soham Ghosh
Measurement validity in Item Response Theory depends on appropriately modeling dependencies between items when these reflect meaningful theoretical structures rather than random me…