Error estimation and reduction with cross correlations

Martin Weigel, Wolfhard Janke

Research output: Contribution to journalArticle

18 Citations (Scopus)

Abstract

Besides the well-known effect of autocorrelations in time series of Monte Carlo simulation data resulting from the underlying Markov process, using the same data pool for computing various estimates entails additional cross correlations. This effect, if not properly taken into account, leads to systematically wrong error estimates for combined quantities. Using a straightforward recipe of data analysis employing the jackknife or similar resampling techniques, such problems can be avoided. In addition, a covariance analysis allows for the formulation of optimal estimators with often significantly reduced variance as compared to more conventional averages.
Original languageEnglish
Article number066701
JournalPhysical Review E
Volume81
DOIs
Publication statusPublished - 8 Jun 2010

Fingerprint

Error Reduction
Error Estimation
Cross-correlation
cross correlation
Markov processes
Jackknife
data simulation
Resampling
estimates
Autocorrelation
estimators
Markov Process
autocorrelation
Error Estimates
Data analysis
Time series
Monte Carlo Simulation
Estimator
formulations
Formulation

Bibliographical note

The full text is not available on the repository.

Cite this

Error estimation and reduction with cross correlations. / Weigel, Martin; Janke, Wolfhard.

In: Physical Review E, Vol. 81, 066701, 08.06.2010.

Research output: Contribution to journalArticle

@article{55140cb5c3304fe9ab589beaeac4a2ac,
title = "Error estimation and reduction with cross correlations",
abstract = "Besides the well-known effect of autocorrelations in time series of Monte Carlo simulation data resulting from the underlying Markov process, using the same data pool for computing various estimates entails additional cross correlations. This effect, if not properly taken into account, leads to systematically wrong error estimates for combined quantities. Using a straightforward recipe of data analysis employing the jackknife or similar resampling techniques, such problems can be avoided. In addition, a covariance analysis allows for the formulation of optimal estimators with often significantly reduced variance as compared to more conventional averages.",
author = "Martin Weigel and Wolfhard Janke",
note = "The full text is not available on the repository.",
year = "2010",
month = "6",
day = "8",
doi = "10.1103/PhysRevE.81.066701",
language = "English",
volume = "81",
journal = "Physical Review E",
issn = "1539-3755",
publisher = "APS",

}

TY - JOUR

T1 - Error estimation and reduction with cross correlations

AU - Weigel, Martin

AU - Janke, Wolfhard

N1 - The full text is not available on the repository.

PY - 2010/6/8

Y1 - 2010/6/8

N2 - Besides the well-known effect of autocorrelations in time series of Monte Carlo simulation data resulting from the underlying Markov process, using the same data pool for computing various estimates entails additional cross correlations. This effect, if not properly taken into account, leads to systematically wrong error estimates for combined quantities. Using a straightforward recipe of data analysis employing the jackknife or similar resampling techniques, such problems can be avoided. In addition, a covariance analysis allows for the formulation of optimal estimators with often significantly reduced variance as compared to more conventional averages.

AB - Besides the well-known effect of autocorrelations in time series of Monte Carlo simulation data resulting from the underlying Markov process, using the same data pool for computing various estimates entails additional cross correlations. This effect, if not properly taken into account, leads to systematically wrong error estimates for combined quantities. Using a straightforward recipe of data analysis employing the jackknife or similar resampling techniques, such problems can be avoided. In addition, a covariance analysis allows for the formulation of optimal estimators with often significantly reduced variance as compared to more conventional averages.

U2 - 10.1103/PhysRevE.81.066701

DO - 10.1103/PhysRevE.81.066701

M3 - Article

VL - 81

JO - Physical Review E

JF - Physical Review E

SN - 1539-3755

M1 - 066701

ER -