Generalized additive models for location, scale and shape for high dimensional data—a flexible approach based on boosting

By Andreas Mayr, Nora Fenske, B. Hofner, Thomas Kneib, Matthias Schmid. 2012

178 citations recorded by the source corpus.

Summary. Generalized additive models for location, scale and shape (GAMLSSs) are a popular semiparametric modelling approach that, in contrast with conventional generalized additive models, regress not only the expected mean but also every distribution parameter (e.g. location, scale and shape) to a set of covariates. Current fitting procedures for GAMLSSs are infeasible for high dimensional data set‐ups and require variable selection based on (potentially problematic) information criteria. The present work describes a boosting algorithm for high dimensional GAMLSSs that was developed to overcome these limitations. Specifically, the new algorithm was designed to allow the simultaneous estimation of predictor effects and variable selection. The algorithm proposed was applied to Munich rental guide data, which are used by landlords and tenants as a reference for the average rent of a flat depending on its characteristics and spatial features. The net rent predictions that resulted from the high dimensional GAMLSSs were found to be highly competitive and covariate‐specific prediction intervals showed a major improvement over classical generalized additive models.

Subfields: Economics and Econometrics

View Paper: https://www.semanticscholar.org/paper/81f9e5db907963acb203bec2fdbed93861ad62f7

How researchers see this paper

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

In Economics and Econometrics

Citations in the corpus

Most Likely Transformations (2015, methodology): “We fitted the GAMLSS models by boosting; the number of boosting iterations was determined via sample splitting (Mayr et al. 2012).”
Twenty years of P-splines (2015, background): “Boosting has many enthusiastic proponents, and its use has been extended to non-normal data and additive models and other smooth structures (Mayr et al., 2012).”
A general framework for functional regression modelling (2017, methodology): “…models (GA(M)Ms) (e.g., Eilers and Marx, 2002; Wood, 2006a; Schmid and Hothorn, 2008; Hothorn et al., 2016; Wood, 2016b), quantile regression (e.g., Koenker, 2005; Fenske et al., 2011) Statistical Modelling 2017; 17(1–2): 1–35 or GAMLSS (e.g., Rigby and Stasinopoulos, 2005; Mayr et al., 2012).”
Geostatistical radar–raingauge merging: A novel method for the quantification of rain estimation accuracy (2014, methodology): “Following [16], a generalized additive model for location, scale and shape (GAMLSS; [31,32]) was fitted to the CV residuals for both the OK (left) and KED (right) estimators.”
Beyond mean regression (2013, methodology): “Such a proposal is made in Mayr et al. (2012) and implemented in the R add-on package gamboostLSS (Hofner et al., 2011).”
Conditional transformation models (2012, methodology): “The computation of prediction intervals based on pairs of such models is rather straightforward (Mayr, Hothorn, and Fenske 2012b).”