Minimum CRPS vs. maximum likelihood

Blog

news

In a new paper in Monthly Weather Review, minimum CRPS and maximum likelihood estimation are compared for fitting heteroscedastic (or nonhomogenous) regression models under different response distributions. Minimum CRPS is more robust to distributional misspecification while maximum likelihood is slightly more efficient under correct specification. An R implementation is available in the crch package. Read more ›

Partially additive (generalized) linear model trees

Blog

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

Blog

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

Blog

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

Blog

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

Blog

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

Blog

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

Blog

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

Blog

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

Blog

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 ›