Infrastructure for Forecasting and Assessment of Probabilistic Models

Unified infrastructure for probabilistic models and distributional regressions: Probabilistic forecasting of in-sample and out-of-sample of probabilities, densities, quantiles, and moments. Probabilistic residuals and scoring via log-score (or log-likelihood), (continuous) ranked probability score, etc. Diagnostic graphics like rootograms, PIT histograms, (randomized) quantile residual Q-Q plots, and reliagrams (reliability diagrams).

Modular object-oriented implementation with support for many model objects, including lm, glm, glm.nb, gamlss, bamlss, hurdle, zeroinfl, zerotrunc, crch, betareg, and more to come.

Rootogram Probabilistic model evaluation with topmodels: Rootogram

PIT histogram Probabilistic model evaluation with topmodels: PIT Histogram

Reliagram Probabilistic model evaluation with topmodels: Reliagram

Q-Q residuals plot Probabilistic model evaluation with topmodels: Q-Q residuals plot