Jove
Visualize
Contact Us
JoVE
x logofacebook logolinkedin logoyoutube logo
ABOUT JoVE
OverviewLeadershipBlogJoVE Help Center
AUTHORS
Publishing ProcessEditorial BoardScope & PoliciesPeer ReviewFAQSubmit
LIBRARIANS
TestimonialsSubscriptionsAccessResourcesLibrary Advisory BoardFAQ
RESEARCH
JoVE JournalMethods CollectionsJoVE Encyclopedia of ExperimentsArchive
EDUCATION
JoVE CoreJoVE BusinessJoVE Science EducationJoVE Lab ManualFaculty Resource CenterFaculty Site
Terms & Conditions of Use
Privacy Policy
Policies

Related Experiment Videos

Nonparametric estimation of covariance structure in longitudinal data

P J Diggle1, A P Verbyla

  • 1Department of Mathematics and Statistics, Lancaster University, U.K.

Biometrics
|June 18, 1998
PubMed
Summary

This study introduces a new nonparametric method for estimating covariance structure in longitudinal data, crucial for analyzing treatment effects over time. This approach, without assuming stationarity, improves analysis for complex, unbalanced datasets.

Related Concept Videos

You might also read

Related Articles

Articles linked to this work by shared authors, journal, and citation graph.

Sort by
Same author

The use of continuous electronic prescribing data to infer trends in antimicrobial consumption and estimate the impact of stewardship interventions in hospitalized children.

The Journal of antimicrobial chemotherapy·2021
Same author

A spatial analysis of giardiasis and cryptosporidiosis in relation to public water supply distribution in North West England.

Spatial and spatio-temporal epidemiology·2018
Same author

Evidence of multiple intraspecific transmission routes for Leptospira acquisition in Norway rats (Rattus norvegicus).

Epidemiology and infection·2017
Same author

A statistical model to describe longitudinal and correlated metabolic risk factors: the Whitehall II prospective study.

Journal of public health (Oxford, England)·2017
Same author

The impact of Type 2 diabetes prevention programmes based on risk-identification and lifestyle intervention intensity strategies: a cost-effectiveness analysis.

Diabetic medicine : a journal of the British Diabetic Association·2017
Same author

Feasibility and acceptability of web-based enhanced relapse prevention for bipolar disorder (ERPonline): trial protocol.

Contemporary clinical trials·2015

Area of Science:

  • Biostatistics
  • Longitudinal Data Analysis
  • Nonparametric Statistics

Background:

  • Longitudinal studies often involve correlated observations within subjects, requiring accounting for covariance structure.
  • Traditional sample covariance matrices are impractical for unbalanced or long-sequence data.
  • Existing nonparametric methods often assume stationarity, limiting their applicability.

Purpose of the Study:

  • To develop a nonparametric estimator for covariance structure that does not assume stationarity.
  • To provide a flexible tool for analyzing complex longitudinal data.
  • To demonstrate the utility of the proposed estimator as a diagnostic tool.

Main Methods:

  • Kernel weighted local linear regression smoothing of sample variogram ordinates.

Related Experiment Videos

  • Smoothing of squared residuals to estimate covariance structure.
  • Application to real-world datasets without stationarity assumptions.

    • Main Results:

    • The proposed method provides a nonparametric estimate of covariance structure without assuming stationarity.
    • Demonstrated effectiveness as a diagnostic tool in analyzing blood pressure and CD4 cell count data.
    • Highlights the estimator's potential for handling unbalanced and long-sequence longitudinal data.

    Conclusions:

    • The developed kernel smoothing technique offers a robust nonparametric approach to estimating covariance structure in longitudinal studies.
    • The method is valuable for diagnostic purposes, especially with non-stationary data.
    • Further research is needed to integrate this estimator into formal statistical inference for mean profiles.