Category: news. Page 5
Partially additive (generalized) linear model trees
news
The PALM tree algorithm for partially additive (generalized) linear model trees is introduced along with the R package palmtree. One potential application is modeling of treatment-subgroup interactions while adjusting for global additive effects. Read more ›
Thunderstorm forecasting with GAMs
news
Boosted binary generalized additive models (GAMs) with stability selection and corresponding MCMC-based credibility intervals are discussed in a new MWR paper as a probabilistic forecasting method for the occurrence of thunderstorms. Read more ›
MPT trees published in BRM
news
Multinomial processing trees are recursively partitioned to capture heterogeneity in latent cognitive processing steps. Accompanied by the R function mpttree in the psychotree package, combining partykit::mob and psychotools::mpt. Read more ›
Clustered covariances in sandwich 2.5-0
news
Version 2.5-0 of the R package 'sandwich' is available from CRAN now with enhanced object-oriented clustered covariances (for lm, glm, survreg, polr, hurdle, zeroinfl, betareg, ...). The software and corresponding vignette have been improved considerably based on helpful and constructive reviewer feedback as well as various bug reports. Read more ›
Evaluation of the 2018 FIFA World Cup forecast
news
A look back the 2018 FIFA World Cup in Russia to check whether our tournament forecast based on the bookmaker consensus model was any good... Read more ›
Sankey diagram for the 2018 FIFA World Cup forecast
news
The probabilistic forecast from the bookmaker consensus model for the 2018 FIFA World Cup is visualized in an interactive Sankey diagram, highlighting the teams' most likely progress through the tournament. Read more ›
Probabilistic forecasting for the 2018 FIFA World Cup
news
Using a consensus model based on quoted bookmakers' odds winning probabilities for all competing teams in the FIFA World Cup are obtained: The favorite is Brazil, closely followed by the defending World Champion Germany. Read more ›
Distributional regression forests on arXiv
news
Distributional regression trees and forests provide flexible data-driven probabilistic forecasts by blending distributional models (for location, scale, shape, and beyond) with regression trees and random forests. Accompanied by the R package disttree. Read more ›
BAMLSS paper published in JCGS
news
Bayesian additive models for location, scale, and shape (and beyond) provide a general framework for distributional regression. Accompanied by the R package bamlss. Read more ›
GLMM trees published in BRM
news
Generalized linear mixed-effects model trees, especially for detecting treatment-subgroup interactions in clustered data. Accompanied by the R package glmertree, combining partykit::glmtree and lme4::glmer. Read more ›