Distributional Regression Forests for Probabilistic Precipitation Forecasting in Complex Terrain

Abstract

To obtain a probabilistic model for a dependent variable based on some set of explanatory variables, a distributional approach is often adopted where the parameters of the distribution are linked to regressors. In many classical models this only captures the location of the distribution but over the last decade there has been increasing interest in distributional regression approaches modeling all parameters including location, scale, and shape. Notably, so-called non-homogeneous Gaussian regression (NGR) models both mean and variance of a Gaussian response and is particularly popular in weather forecasting. Moreover, generalized additive models for location, scale, and shape (GAMLSS) provide a framework where each distribution parameter is modeled separately capturing smooth linear or nonlinear effects. However, when variable selection is required and/or there are non-smooth dependencies or interactions (especially unknown or of high-order), it is challenging to establish a good GAMLSS. A natural alternative in these situations would be the application of regression trees or random forests but, so far, no general distributional framework is available for these. Therefore, a framework for distributional regression trees and forests is proposed that blends regression trees and random forests with classical distributions from the GAMLSS framework as well as their censored or truncated counterparts. To illustrate these novel approaches in practice, they are employed to obtain probabilistic precipitation forecasts at numerous sites in a mountainous region based on a large number of numerical weather prediction quantities. It is shown that the novel distributional regression forests automatically select variables and interactions, performing on par or often even better than GAMLSS specified either through prior meteorological knowledge or a computationally more demanding boosting approach.

References in corpus (3)

• Neural networks for post-processing ensemble weather forecasts
• Bayesian structured additive distributional regression with an application to regional income inequality in Germany
• Censored and shifted gamma distribution based EMOS model for probabilistic quantitative precipitation forecasting

Cited by in corpus (16)

• Neural networks for post-processing ensemble weather forecasts
• Statistical Postprocessing for Weather Forecasts -- Review, Challenges and Avenues in a Big Data World
• A review of predictive uncertainty estimation with machine learning
• Isotonic Distributional Regression
• XGBoostLSS -- An extension of XGBoost to probabilistic forecasting
• Interpretable AI-Driven Discovery of Terrain-Precipitation Relationships for Enhanced Climate Insights
• CatBoostLSS -- An extension of CatBoost to probabilistic forecasting
• Quantile based modelling of diurnal temperature range with the five-parameter lambda distribution
• What Makes Forest-Based Heterogeneous Treatment Effect Estimators Work?
• Random Forests for dependent data
• A review of machine learning concepts and methods for addressing challenges in probabilistic hydrological post-processing and forecasting
• The Power of Unbiased Recursive Partitioning: A Unifying View of CTree, MOB, and GUIDE
• Circular Regression Trees and Forests with an Application to Probabilistic Wind Direction Forecasting
• Combinations of distributional regression algorithms with application in uncertainty estimation of corrected satellite precipitation products
• A Random Forest Approach for Modeling Bounded Outcomes
• Multivariate Uncertainty Quantification with Tomographic Quantile Forests