Model-Based Recursive Partitioning

By Achim Zeileis, Torsten Hothorn, Kurt Hornik. 2008

741 citations recorded by the source corpus.

Subfields: Artificial Intelligence, Finance, Statistics and Probability, Optimization & Training

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Artificial Intelligence

In Finance

In Statistics and Probability

In Optimization & Training

Citations in the corpus

Machine Learning Methods That Economists Should Know About (2019, methodology): “The approach of estimating simple models in the leaves of shallow trees can be applied to other types of models; see [Zeileis et al., 2008] for an early version of this idea, although that paper did not provide theoretical guarantees or confidence intervals.”
Invariant Causal Prediction for Nonlinear Models (2017, neutral): “rroni correction. See Appendix B.6 for further details. Another interesting possibility for future work would be to devise a conditional independence test based on model-based recursive partitioning (Zeileis et al., 2008; Hothorn and Zeileis, 2015). Non-trivial, assumption-free conditional independe”
Local Linear Forests (2018, background): “Zeileis, Hothorn, and Hornik (2008), and later Rusch and Zeileis (2013), proposed not only prediction, but recursive partitioning via fitting a separate model in each leaf, similar to the residual splitting strategy of local linear forests.”
Recursive partitioning for heterogeneous causal effects (2015, background): “(19) propose using statistical tests based on improvements in goodness of fit to determine when to stop growing the tree, rather than relying on cross-validation, but for ease of comparison with CART, in this paper we will stay closer to traditional CART in terms of growing deep trees and pruning th”
Fairer and more accurate, but for whom? (2017, methodology): “Our partitioning procedure follows the approach of [15, 22] for model-based recursive binary partitioning, and relies on a modied version of the corresponding soware.”
Causal Falling Rule Lists (2015, background): “(Su and Nickerson 2009; Su, Wang, and Fan 2004; Zeileis, Hothorn, and Hornik 2008) use a model-based node splitting criteria in building a tree that identifies heterogeniety of some quantity of interest based on covariates.”
Accounting for Individual Differences in Bradley-Terry Models by Means of Recursive Partitioning (2011, methodology): “A particularly convenient one—due to its generality and ease of computation—is the usage of fluctuation tests (Zeileis & Hornik, 2007) as adopted in the model-based recursive partitioning framework of Zeileis, Hothorn, and Hornik (2008).”
The Impact of Machine Learning on Economics (2018, methodology): “This paper builds on earlier work on “model-based recursive partitioning (Zeileis et al., 2008), which looked at recursive partitioning for more complex models, but did not provide statistical properties.”
Machine Learning Methods for Estimating Heterogeneous Causal Eects (2015, methodology): “[36] diﬀer in that they base the split in sample on model ﬁt, and in that they do not consider cross-validation for selecting a complexity tuning parameter, so the issue of selecting an out-of-sample goodness of ﬁt metric does not arise.”
Random Forests for dependent data (2020, methodology): “Generalizations of the loss function used in RF have been shown to be successful in many different contexts like quantile estimation [47], distributional forests [61], causal forests [4], manifold forests emulating convolutional neural networks [55], and others [40, 73, 14, 58, 68].”
Interpretable Machine Learning - A Brief History, State-of-the-Art and Challenges (2020, background): “e, to interpret deep neural networks or tree ensembles. Regression analysis and rule-based ML remain important and active research areas to this day and are blending together (e.g., model-based trees [128], RuleFit [33]). Many extensions of the linear regression model exist [45,25,38] and new extens”
Applied Predictive Modeling (2013, methodology): “Also, several techniques exist (Frank et al. 1998; Loh 2002; Chan and Loh 2004; Zeileis et al. 2008) that use more complex models in the terminal nodes, similar to M5 and Cubist.”
Distributional regression forests for probabilistic precipitation forecasting in complex terrain (2018, methodology): “A different framework to build a likelihood-based tree is provided by the MOB algorithm which is based on M-fluctuation tests (Zeileis et al. 2008).”