Bayesian effect fusion for categorical predictors
From MaRDI portal
Abstract: In this paper, we propose a Bayesian approach to obtain a sparse representation of the effect of a categorical predictor in regression type models. As the effect of a categorical predictor is captured by a group of level effects, sparsity cannot only be achieved by excluding single irrelevant level effects but also by excluding the whole group of effects associated to a predictor or by fusing levels which have essentially the same effect on the response. To achieve this goal, we propose a prior which allows for almost perfect as well as almost zero dependence between level effects a priori. We show how this prior can be obtained by specifying spike and slab prior distributions on all effect differences associated to one categorical predictor and how restricted fusion can be implemented. An efficient MCMC method for posterior computation is developed. The performance of the proposed method is investigated on simulated data. Finally, we illustrate its application on real data from EU-SILC.
Recommendations
- Regularization and model selection with categorial effect modifiers
- Sparse modeling of categorial explanatory variables
- Bayesian effect selection in structured additive distributional regression models
- Bayesian linear regression with sparse priors
- Tree-structured modelling of categorical predictors in generalized additive regression
Cites work
- scientific article; zbMATH DE number 1034042 (Why is no real title available?)
- scientific article; zbMATH DE number 845714 (Why is no real title available?)
- Bayesian Variable Selection in Linear Regression
- Bayesian cluster analysis
- Bayesian effect fusion for categorical predictors
- Bayesian regularization via graph Laplacian
- Bayesian variable selection with related predictors
- Effect fusion using model-based clustering
- Improved criteria for clustering based on the posterior similarity matrix
- Inference with normal-gamma prior distributions in regression problems
- Model Determination for Categorical Data With Factor Level Merging
- Model Selection and Estimation in Regression with Grouped Variables
- Penalized regression, standard errors, and Bayesian Lassos
- Propriety of posteriors with improper priors in hierachical linear mixed models.
- Regularization and Variable Selection Via the Elastic Net
- Simultaneous Factor Selection and Collapsing Levels in ANOVA
- Sparse modeling of categorial explanatory variables
- Sparsity and Smoothness Via the Fused Lasso
- Spike and slab variable selection: frequentist and Bayesian strategies
- Spike-and-slab priors for function selection in structured additive regression models
- The Bayesian Lasso
- Tree-structured modelling of categorical predictors in generalized additive regression
Cited in
(10)- The Lasso on latent indices for regression modeling with ordinal categorical predictors
- Calibration procedures for approximate Bayesian credible sets
- Variable Selection in the Presence of Factors: A Model Selection Perspective
- Bayesian effect fusion for categorical predictors
- Regularization and model selection with categorial effect modifiers
- A Bayesian mixture model for changepoint estimation using ordinal predictors
- Sparse modeling of categorial explanatory variables
- Handling categorical features with many levels using a product partition model
- Effect fusion using model-based clustering
- Random effects clustering in multilevel modeling: choosing a proper partition
This page was built for publication: Bayesian effect fusion for categorical predictors
Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q1738136)
