A differential-geometric approach to generalized linear models with grouped predictors

Risultato della ricerca: Article

2 Citazioni (Scopus)

Abstract

We propose an extension of the differential-geometric least angle regression method to per- form sparse group inference in a generalized linear model. An efficient algorithm is proposed to compute the solution curve. The proposed group differential-geometric least angle regression method has important properties that distinguish it from the group lasso. First, its solution curve is based on the invariance properties of a generalized linear model. Second, it adds groups of variables based on a group equiangularity condition, which is shown to be related to score statis- tics. An adaptive version, which includes weights based on the Kullback–Leibler divergence, improves its variable selection features and is shown to have oracle properties when the number of predictors is fixed.
Lingua originaleEnglish
pagine (da-a)563-577
Numero di pagine15
RivistaBiometrika
Volume103
Stato di pubblicazionePublished - 2016

Fingerprint

Geometric Approach
Generalized Linear Model
Predictors
Linear Models
linear models
Invariance
Feature extraction
statistics
Statistics
Weights and Measures
Regression
Oracle Property
Angle
Score Statistic
methodology
Kullback-Leibler Divergence
Curve
Lasso
Variable Selection
Efficient Algorithms

All Science Journal Classification (ASJC) codes

  • Statistics and Probability
  • Mathematics(all)
  • Agricultural and Biological Sciences (miscellaneous)
  • Agricultural and Biological Sciences(all)
  • Statistics, Probability and Uncertainty
  • Applied Mathematics

Cita questo

A differential-geometric approach to generalized linear models with grouped predictors. / Mineo, Angelo; Augugliaro, Luigi; Wit, Ernst C.

In: Biometrika, Vol. 103, 2016, pag. 563-577.

Risultato della ricerca: Article

@article{881d2f0db4c34922a7b3159fb454798c,
title = "A differential-geometric approach to generalized linear models with grouped predictors",
abstract = "We propose an extension of the differential-geometric least angle regression method to per- form sparse group inference in a generalized linear model. An efficient algorithm is proposed to compute the solution curve. The proposed group differential-geometric least angle regression method has important properties that distinguish it from the group lasso. First, its solution curve is based on the invariance properties of a generalized linear model. Second, it adds groups of variables based on a group equiangularity condition, which is shown to be related to score statis- tics. An adaptive version, which includes weights based on the Kullback–Leibler divergence, improves its variable selection features and is shown to have oracle properties when the number of predictors is fixed.",
author = "Angelo Mineo and Luigi Augugliaro and Wit, {Ernst C.}",
year = "2016",
language = "English",
volume = "103",
pages = "563--577",
journal = "Biometrika",
issn = "0006-3444",
publisher = "Oxford University Press",

}

TY - JOUR

T1 - A differential-geometric approach to generalized linear models with grouped predictors

AU - Mineo, Angelo

AU - Augugliaro, Luigi

AU - Wit, Ernst C.

PY - 2016

Y1 - 2016

N2 - We propose an extension of the differential-geometric least angle regression method to per- form sparse group inference in a generalized linear model. An efficient algorithm is proposed to compute the solution curve. The proposed group differential-geometric least angle regression method has important properties that distinguish it from the group lasso. First, its solution curve is based on the invariance properties of a generalized linear model. Second, it adds groups of variables based on a group equiangularity condition, which is shown to be related to score statis- tics. An adaptive version, which includes weights based on the Kullback–Leibler divergence, improves its variable selection features and is shown to have oracle properties when the number of predictors is fixed.

AB - We propose an extension of the differential-geometric least angle regression method to per- form sparse group inference in a generalized linear model. An efficient algorithm is proposed to compute the solution curve. The proposed group differential-geometric least angle regression method has important properties that distinguish it from the group lasso. First, its solution curve is based on the invariance properties of a generalized linear model. Second, it adds groups of variables based on a group equiangularity condition, which is shown to be related to score statis- tics. An adaptive version, which includes weights based on the Kullback–Leibler divergence, improves its variable selection features and is shown to have oracle properties when the number of predictors is fixed.

UR - http://hdl.handle.net/10447/193784

UR - http://biomet.oxfordjournals.org/content/103/3/563.abstract

M3 - Article

VL - 103

SP - 563

EP - 577

JO - Biometrika

JF - Biometrika

SN - 0006-3444

ER -