A note on estimating the number of superimposed exponential signals by the cross-validation approach

Y. Wu, K. W. Tarn, Mei-Mei Zen, F. Li

Research output: Contribution to journalConference article

Abstract

In this paper, a procedure based on the delete-1 cross-validation is given for estimating the number of superimposed exponential signals, its limiting behavior is explored and it is shown that the probability of overestimating the true number of signals is greater than a positive constant for sufficiently large samples. Also a general procedure based on the cross-validation is presented when the deletion proceeds according to a collection of subsets of indices. The result is similar to the delete-1 cross-validation if the number of deletions is fixed. The simulation results are provided for the performance of the procedure when the collections of subsets of indices are chosen as those suggested by Shao (1) in a linear model selection problem.

Original languageEnglish
Pages (from-to)1579-1589
Number of pages11
JournalCommunications in Statistics - Theory and Methods
Volume30
Issue number8-9
DOIs
Publication statusPublished - 2001 Jan 1
EventStatistics Reflections on the Past and Visions for the Future-International Conference in Honor of Prof. C.R. Rao on the Occasion of His 80th Birthday- - San Antonio, TX, United States
Duration: 2000 Mar 162000 Mar 19

Fingerprint

Cross-validation
Deletion
Subset
Limiting Behavior
Model Selection
Linear Model
Simulation

All Science Journal Classification (ASJC) codes

  • Statistics and Probability

Cite this

@article{7d554dee45194108b2f25272102d4c94,
title = "A note on estimating the number of superimposed exponential signals by the cross-validation approach",
abstract = "In this paper, a procedure based on the delete-1 cross-validation is given for estimating the number of superimposed exponential signals, its limiting behavior is explored and it is shown that the probability of overestimating the true number of signals is greater than a positive constant for sufficiently large samples. Also a general procedure based on the cross-validation is presented when the deletion proceeds according to a collection of subsets of indices. The result is similar to the delete-1 cross-validation if the number of deletions is fixed. The simulation results are provided for the performance of the procedure when the collections of subsets of indices are chosen as those suggested by Shao (1) in a linear model selection problem.",
author = "Y. Wu and Tarn, {K. W.} and Mei-Mei Zen and F. Li",
year = "2001",
month = "1",
day = "1",
doi = "10.1081/STA-100105685",
language = "English",
volume = "30",
pages = "1579--1589",
journal = "Communications in Statistics - Theory and Methods",
issn = "0361-0926",
publisher = "Taylor and Francis Ltd.",
number = "8-9",

}

A note on estimating the number of superimposed exponential signals by the cross-validation approach. / Wu, Y.; Tarn, K. W.; Zen, Mei-Mei; Li, F.

In: Communications in Statistics - Theory and Methods, Vol. 30, No. 8-9, 01.01.2001, p. 1579-1589.

Research output: Contribution to journalConference article

TY - JOUR

T1 - A note on estimating the number of superimposed exponential signals by the cross-validation approach

AU - Wu, Y.

AU - Tarn, K. W.

AU - Zen, Mei-Mei

AU - Li, F.

PY - 2001/1/1

Y1 - 2001/1/1

N2 - In this paper, a procedure based on the delete-1 cross-validation is given for estimating the number of superimposed exponential signals, its limiting behavior is explored and it is shown that the probability of overestimating the true number of signals is greater than a positive constant for sufficiently large samples. Also a general procedure based on the cross-validation is presented when the deletion proceeds according to a collection of subsets of indices. The result is similar to the delete-1 cross-validation if the number of deletions is fixed. The simulation results are provided for the performance of the procedure when the collections of subsets of indices are chosen as those suggested by Shao (1) in a linear model selection problem.

AB - In this paper, a procedure based on the delete-1 cross-validation is given for estimating the number of superimposed exponential signals, its limiting behavior is explored and it is shown that the probability of overestimating the true number of signals is greater than a positive constant for sufficiently large samples. Also a general procedure based on the cross-validation is presented when the deletion proceeds according to a collection of subsets of indices. The result is similar to the delete-1 cross-validation if the number of deletions is fixed. The simulation results are provided for the performance of the procedure when the collections of subsets of indices are chosen as those suggested by Shao (1) in a linear model selection problem.

UR - http://www.scopus.com/inward/record.url?scp=33744852562&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=33744852562&partnerID=8YFLogxK

U2 - 10.1081/STA-100105685

DO - 10.1081/STA-100105685

M3 - Conference article

AN - SCOPUS:33744852562

VL - 30

SP - 1579

EP - 1589

JO - Communications in Statistics - Theory and Methods

JF - Communications in Statistics - Theory and Methods

SN - 0361-0926

IS - 8-9

ER -