A method for choosing the smoothing parameter in a semi-parametric model for detecting change-points in blood flow

Sung Won Han, Rickson C. Mesquita, Theresa M. Busch, Mary E. Putt

Research output: Contribution to journalArticle

6 Citations (Scopus)

Abstract

In a smoothing spline model with unknown change-points, the choice of the smoothing parameter strongly influences the estimation of the change-point locations and the function at the change-points. In a tumor biology example, where change-points in blood flow in response to treatment were of interest, choosing the smoothing parameter based on minimizing generalized cross-validation (GCV) gave unsatisfactory estimates of the change-points. We propose a new method, aGCV, that re-weights the residual sum of squares and generalized degrees of freedom terms from GCV. The weight is chosen to maximize the decrease in the generalized degrees of freedom as a function of the weight value, while simultaneously minimizing aGCV as a function of the smoothing parameter and the change-points. Compared with GCV, simulation studies suggest that the aGCV method yields improved estimates of the change-point and the value of the function at the change-point.

Original languageEnglish
Pages (from-to)26-45
Number of pages20
JournalJournal of Applied Statistics
Volume41
Issue number1
DOIs
Publication statusPublished - 2014 Jan 1
Externally publishedYes

Fingerprint

Smoothing Parameter
Semiparametric Model
Change Point
Blood Flow
Generalized Cross-validation
Degree of freedom
Point Location
Smoothing Splines
Smoothing
Semiparametric model
Change point
Sum of squares
Estimate
Biology
Tumor
Maximise
Simulation Study
Unknown
Decrease
Term

Keywords

  • change-points
  • generalized cross-validation
  • generalized degrees of freedom
  • partial spline
  • smoothing spline

ASJC Scopus subject areas

  • Statistics and Probability
  • Statistics, Probability and Uncertainty

Cite this

A method for choosing the smoothing parameter in a semi-parametric model for detecting change-points in blood flow. / Han, Sung Won; Mesquita, Rickson C.; Busch, Theresa M.; Putt, Mary E.

In: Journal of Applied Statistics, Vol. 41, No. 1, 01.01.2014, p. 26-45.

Research output: Contribution to journalArticle

@article{e454ad3c571d4cfbb38ce7e7aa2958e3,
title = "A method for choosing the smoothing parameter in a semi-parametric model for detecting change-points in blood flow",
abstract = "In a smoothing spline model with unknown change-points, the choice of the smoothing parameter strongly influences the estimation of the change-point locations and the function at the change-points. In a tumor biology example, where change-points in blood flow in response to treatment were of interest, choosing the smoothing parameter based on minimizing generalized cross-validation (GCV) gave unsatisfactory estimates of the change-points. We propose a new method, aGCV, that re-weights the residual sum of squares and generalized degrees of freedom terms from GCV. The weight is chosen to maximize the decrease in the generalized degrees of freedom as a function of the weight value, while simultaneously minimizing aGCV as a function of the smoothing parameter and the change-points. Compared with GCV, simulation studies suggest that the aGCV method yields improved estimates of the change-point and the value of the function at the change-point.",
keywords = "change-points, generalized cross-validation, generalized degrees of freedom, partial spline, smoothing spline",
author = "Han, {Sung Won} and Mesquita, {Rickson C.} and Busch, {Theresa M.} and Putt, {Mary E.}",
year = "2014",
month = "1",
day = "1",
doi = "10.1080/02664763.2013.830085",
language = "English",
volume = "41",
pages = "26--45",
journal = "Journal of Applied Statistics",
issn = "0266-4763",
publisher = "Routledge",
number = "1",

}

TY - JOUR

T1 - A method for choosing the smoothing parameter in a semi-parametric model for detecting change-points in blood flow

AU - Han, Sung Won

AU - Mesquita, Rickson C.

AU - Busch, Theresa M.

AU - Putt, Mary E.

PY - 2014/1/1

Y1 - 2014/1/1

N2 - In a smoothing spline model with unknown change-points, the choice of the smoothing parameter strongly influences the estimation of the change-point locations and the function at the change-points. In a tumor biology example, where change-points in blood flow in response to treatment were of interest, choosing the smoothing parameter based on minimizing generalized cross-validation (GCV) gave unsatisfactory estimates of the change-points. We propose a new method, aGCV, that re-weights the residual sum of squares and generalized degrees of freedom terms from GCV. The weight is chosen to maximize the decrease in the generalized degrees of freedom as a function of the weight value, while simultaneously minimizing aGCV as a function of the smoothing parameter and the change-points. Compared with GCV, simulation studies suggest that the aGCV method yields improved estimates of the change-point and the value of the function at the change-point.

AB - In a smoothing spline model with unknown change-points, the choice of the smoothing parameter strongly influences the estimation of the change-point locations and the function at the change-points. In a tumor biology example, where change-points in blood flow in response to treatment were of interest, choosing the smoothing parameter based on minimizing generalized cross-validation (GCV) gave unsatisfactory estimates of the change-points. We propose a new method, aGCV, that re-weights the residual sum of squares and generalized degrees of freedom terms from GCV. The weight is chosen to maximize the decrease in the generalized degrees of freedom as a function of the weight value, while simultaneously minimizing aGCV as a function of the smoothing parameter and the change-points. Compared with GCV, simulation studies suggest that the aGCV method yields improved estimates of the change-point and the value of the function at the change-point.

KW - change-points

KW - generalized cross-validation

KW - generalized degrees of freedom

KW - partial spline

KW - smoothing spline

UR - http://www.scopus.com/inward/record.url?scp=84889089844&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84889089844&partnerID=8YFLogxK

U2 - 10.1080/02664763.2013.830085

DO - 10.1080/02664763.2013.830085

M3 - Article

AN - SCOPUS:84889089844

VL - 41

SP - 26

EP - 45

JO - Journal of Applied Statistics

JF - Journal of Applied Statistics

SN - 0266-4763

IS - 1

ER -