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

Bias in slope estimates for the linear errors in variables model by the variance ratio method

S D Edland1

  • 1Department of Environmental Health, University of Washington, Seattle 98195-4790, USA.

Biometrics
|March 1, 1996
PubMed
Summary

Slope estimates in linear measurement error models are biased when relationships aren't deterministic. This study introduces a model to characterize bias from "errors in equations," highlighting its significant impact.

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

An Update on Career Trajectories and Professional Development Plans of MADURA Undergraduate Mentorship Program Alumni and Continuing Trainees.

HSOA journal of gerontology & geriatric medicine·2025
Same author

Re: Letter to the Editor of Public Health in response to 'The relationships among physical activity, sedentary behaviour, obesity, and quitting behaviours within a cohort of smokers in California'.

Public health·2018
Same author

The relationships among physical activity, sedentary behaviour, obesity and quitting behaviours within a cohort of smokers in California.

Public health·2016
Same author

Donepezil delays progression to AD in MCI subjects with depressive symptoms.

Neurology·2009
Same author

Incident dementia in women is preceded by weight loss by at least a decade.

Neurology·2007
Same author

Coronary artery bypass grafting is not a risk factor for dementia or Alzheimer disease.

Neurology·2005

Area of Science:

  • Statistics
  • Econometrics

Background:

  • Linear measurement error models assume known error variance ratios.
  • Bias arises when the linear relationship is not strictly law-like.

Purpose of the Study:

  • To describe an eight-parameter linear measurement error model.
  • To characterize the asymptotic bias of slope estimates when "errors in equations" are present.

Main Methods:

  • Developed a general applicability linear measurement error model.
  • Incorporated an optional "errors in equations" term for bias analysis.

Main Results:

  • Slope estimates are biased if the linear relationship is not deterministic.
  • The proposed model explicitly characterizes this asymptotic bias.

Related Experiment Videos

Conclusions:

  • Bias in slope estimates can be substantial.
  • Recognizing "errors in equations" is crucial for accurate measurement error modeling.