Sparse graphical models via calibrated concave convex procedure with application to fMRI data

Sungtaek Son, Cheolwoo Park, Yongho Jeon

Research output: Contribution to journalArticle

Abstract

This paper proposes a calibrated concave convex procedure (calibrated CCCP) for high-dimensional graphical model selection. The calibrated CCCP approach for the smoothly clipped absolute deviation (SCAD) penalty is known to be path-consistent with probability converging to one in linear regression models. We implement the calibrated CCCP method with the SCAD penalty for the graphical model selection. We use a quadratic objective function for undirected Gaussian graphical models and adopt the SCAD penalty for sparse estimation. For the tuning procedure, we propose to use columnwise tuning on the quadratic objective function adjusted for test data. In a simulation study, we compare the performance of the proposed method with two existing graphical model estimators for high-dimensional data in terms of matrix error norms and support recovery rate. We also compare the bias and the variance of the estimated matrices. Then, we apply the method to functional magnetic resonance imaging (fMRI) data of an attention deficit hyperactivity disorders (ADHD) patient.

Original languageEnglish
JournalJournal of Applied Statistics
DOIs
Publication statusAccepted/In press - 2019 Jan 1

Fingerprint

Functional Magnetic Resonance Imaging
Graphical Models
Penalty
Deviation
Quadratic Function
Model Selection
Tuning
Objective function
Gaussian Model
High-dimensional Data
Linear Regression Model
Disorder
High-dimensional
Recovery
Simulation Study
Estimator
Norm
Path
Graphical models
Functional magnetic resonance imaging

All Science Journal Classification (ASJC) codes

  • Statistics and Probability
  • Statistics, Probability and Uncertainty

Cite this

@article{aacf1cb8402b478d97aa5f8a35d54880,
title = "Sparse graphical models via calibrated concave convex procedure with application to fMRI data",
abstract = "This paper proposes a calibrated concave convex procedure (calibrated CCCP) for high-dimensional graphical model selection. The calibrated CCCP approach for the smoothly clipped absolute deviation (SCAD) penalty is known to be path-consistent with probability converging to one in linear regression models. We implement the calibrated CCCP method with the SCAD penalty for the graphical model selection. We use a quadratic objective function for undirected Gaussian graphical models and adopt the SCAD penalty for sparse estimation. For the tuning procedure, we propose to use columnwise tuning on the quadratic objective function adjusted for test data. In a simulation study, we compare the performance of the proposed method with two existing graphical model estimators for high-dimensional data in terms of matrix error norms and support recovery rate. We also compare the bias and the variance of the estimated matrices. Then, we apply the method to functional magnetic resonance imaging (fMRI) data of an attention deficit hyperactivity disorders (ADHD) patient.",
author = "Sungtaek Son and Cheolwoo Park and Yongho Jeon",
year = "2019",
month = "1",
day = "1",
doi = "10.1080/02664763.2019.1663158",
language = "English",
journal = "Journal of Applied Statistics",
issn = "0266-4763",
publisher = "Routledge",

}

Sparse graphical models via calibrated concave convex procedure with application to fMRI data. / Son, Sungtaek; Park, Cheolwoo; Jeon, Yongho.

In: Journal of Applied Statistics, 01.01.2019.

Research output: Contribution to journalArticle

TY - JOUR

T1 - Sparse graphical models via calibrated concave convex procedure with application to fMRI data

AU - Son, Sungtaek

AU - Park, Cheolwoo

AU - Jeon, Yongho

PY - 2019/1/1

Y1 - 2019/1/1

N2 - This paper proposes a calibrated concave convex procedure (calibrated CCCP) for high-dimensional graphical model selection. The calibrated CCCP approach for the smoothly clipped absolute deviation (SCAD) penalty is known to be path-consistent with probability converging to one in linear regression models. We implement the calibrated CCCP method with the SCAD penalty for the graphical model selection. We use a quadratic objective function for undirected Gaussian graphical models and adopt the SCAD penalty for sparse estimation. For the tuning procedure, we propose to use columnwise tuning on the quadratic objective function adjusted for test data. In a simulation study, we compare the performance of the proposed method with two existing graphical model estimators for high-dimensional data in terms of matrix error norms and support recovery rate. We also compare the bias and the variance of the estimated matrices. Then, we apply the method to functional magnetic resonance imaging (fMRI) data of an attention deficit hyperactivity disorders (ADHD) patient.

AB - This paper proposes a calibrated concave convex procedure (calibrated CCCP) for high-dimensional graphical model selection. The calibrated CCCP approach for the smoothly clipped absolute deviation (SCAD) penalty is known to be path-consistent with probability converging to one in linear regression models. We implement the calibrated CCCP method with the SCAD penalty for the graphical model selection. We use a quadratic objective function for undirected Gaussian graphical models and adopt the SCAD penalty for sparse estimation. For the tuning procedure, we propose to use columnwise tuning on the quadratic objective function adjusted for test data. In a simulation study, we compare the performance of the proposed method with two existing graphical model estimators for high-dimensional data in terms of matrix error norms and support recovery rate. We also compare the bias and the variance of the estimated matrices. Then, we apply the method to functional magnetic resonance imaging (fMRI) data of an attention deficit hyperactivity disorders (ADHD) patient.

UR - http://www.scopus.com/inward/record.url?scp=85073775298&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85073775298&partnerID=8YFLogxK

U2 - 10.1080/02664763.2019.1663158

DO - 10.1080/02664763.2019.1663158

M3 - Article

AN - SCOPUS:85073775298

JO - Journal of Applied Statistics

JF - Journal of Applied Statistics

SN - 0266-4763

ER -