Sparse modeling of categorial explanatory variables

DOI10.1214/10-AOAS355zbMATH Open1220.62092arXiv1101.1421MaRDI QIDQ542985FDOQ542985

Authors: Jan Gertheiss, Gerhard Tutz 

Publication date: 20 June 2011

Published in: The Annals of Applied Statistics (Search for Journal in Brave)

Abstract: Shrinking methods in regression analysis are usually designed for metric predictors. In this article, however, shrinkage methods for categorial predictors are proposed. As an application we consider data from the Munich rent standard, where, for example, urban districts are treated as a categorial predictor. If independent variables are categorial, some modifications to usual shrinking procedures are necessary. Two $L_1$-penalty based methods for factor selection and clustering of categories are presented and investigated. The first approach is designed for nominal scale levels, the second one for ordinal predictors. Besides applying them to the Munich rent standard, methods are illustrated and compared in simulation studies.

Full work available at URL: https://arxiv.org/abs/1101.1421

Recommendations

• Regularization and model selection with categorial effect modifiers
• Bayesian effect fusion for categorical predictors
• Variable selection in general multinomial logit models
• Shrinkage for categorical regressors
• Regularization and variable selection in categorical regression analyses.

zbMATH Keywords

fused lassocategorial predictorsordinal predictorsrent standardvariable fusion

Mathematics Subject Classification ID

Classification and discrimination; cluster analysis (statistical aspects) (62H30) Applications of statistics to economics (62P20) Ridge regression; shrinkage estimators (Lasso) (62J07) Applications of statistics to social sciences (62P25) Applications of mathematical programming (90C90)

Cites Work

• Simultaneous Factor Selection and Collapsing Levels in ANOVA
• The Adaptive Lasso and Its Oracle Properties
• Least angle regression. (With discussion)
• Variable Selection via Nonconcave Penalized Likelihood and its Oracle Properties
• Title not available (Why is that?)
• The Dantzig selector: statistical estimation when \(p\) is much larger than \(n\). (With discussions and rejoinder).
• Sparsity and Smoothness Via the Fused Lasso
• Regularization and Variable Selection Via the Elastic Net
• Model Selection and Estimation in Regression with Grouped Variables
• The Group Lasso for Logistic Regression
• Relaxed Lasso
• Boosting With theL₂Loss

Cited In (34)

• Modelling Clustered Heterogeneity: Fixed Effects, Random Effects and Mixtures
• Sparsity with sign-coherent groups of variables via the cooperative-Lasso
• Factor selection and structural identification in the interaction ANOVA model
• Bayesian effect fusion for categorical predictors
• Selection and Fusion of Categorical Predictors with L0-Type Penalties
• Lasso-type penalization in the framework of generalized additive models for location, scale and shape
• Extended ordered paired comparison models with application to football data from German Bundesliga
• Fused least absolute shrinkage and selection operator for credit scoring
• A uniform framework for the combination of penalties in generalized structured models
• Supervised dimension reduction for ordinal predictors
• Sparse regression with multi-type regularized feature modeling
• Sparse regression for large data sets with outliers
• Pattern recovery and signal denoising by SLOPE when the design matrix is orthogonal
• Rating scales as predictors -- the old question of scale level and some answers
• Shrinkage for categorical regressors
• Regularization and model selection with categorial effect modifiers
• Structure estimation of binary graphical models on stratified data: application to the description of injury tables for victims of road accidents
• Homogeneity detection for the high-dimensional generalized linear model
• Delete or merge regressors for linear model selection
• Tree-structured modelling of categorical predictors in generalized additive regression
• Tree-Structured Clustering in Fixed Effects Models
• Variable selection and estimation of quantile regression model for stratified data
• Group variable selection in cardiopulmonary cerebral resuscitation data for veterinary patients
• Subject-specific modelling of paired comparison data: A lasso-type penalty approach
• Testing linearity and relevance of ordinal predictors
• Regularization and model selection with categorical predictors and effect modifiers in generalized linear models
• Handling categorical features with many levels using a product partition model
• Effect fusion using model-based clustering
• A regularized hidden Markov model for analyzing the `hot shoe' in football
• A note on coding and standardization of categorical variables in (sparse) group Lasso regression
• Motor insurance claim modelling with factor collapsing and Bayesian model averaging
• The ranking lasso and its application to sport tournaments
• Sparse least trimmed squares regression for analyzing high-dimensional large data sets
• Grouped rank centrality: ranking and grouping from pairwise comparisons simultaneously

Uses Software

• R
• BayesX
• Kernlab
• grplasso
• CasANOVA

This page was built for publication: Sparse modeling of categorial explanatory variables

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q542985)