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

Parametric modelling of cost data: some simulation evidence.

Andrew Briggs1, Richard Nixon, Simon Dixon

  • 1Department of Public Health, Health Economics Research Center, University of Oxford, UK. andrew.briggs@dphpc.ox.ac.uk

Health Economics
|February 3, 2005
PubMed
Summary
This summary is machine-generated.

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

Assessment of Stage Two Hypertension Treatment Plans Written by Generative AI.

Journal of clinical medicine·2026
Same author

A budget apart: the case for ringfencing medicines in the UK.

BMJ (Clinical research ed.)·2026
Same author

Investigating prognostic classifications of preexisting multiple long-term conditions for health outcomes 1 year after COVID-19 hospitalization: A UK prospective observational study.

International journal of infectious diseases : IJID : official publication of the International Society for Infectious Diseases·2026
Same author

Autism and attachment disorder symptoms in the general population: Prevalence, overlap, and burden.

Developmental child welfare·2026
Same author

Methodological Challenges of Emulating a Target Trial to Assess Effectiveness of Timing of PCSK9 Inhibitor Treatment Initiation Post Myocardial Infarction.

Pharmacoepidemiology and drug safety·2026
Same author

Making Radiation Visible: Real-Time Detection of Scatter Radiation Leaks While Using Enhanced Radiation Protection Systems.

JACC. Cardiovascular interventions·2026
Same journal

Lead in Drinking Water and Child Health: Evidence From Jackson, Mississippi.

Health economics·2026
Same journal

Health on the Move: The Impact of Poverty Alleviation Relocation on Healthcare Utilization in China.

Health economics·2026
Same journal

The Effects of Compulsory Licensing: A Case Study of HIV Drugs.

Health economics·2026
Same journal

Beyond Tobacco Prevention: The Effects of Tobacco 21 Laws on Young Adults' Body Weight.

Health economics·2026
Same journal

Assessing the Estimands and Estimates of Hospitalization Rates in Health Economics and Clinical Medicine.

Health economics·2026
Same journal

The Impact of Unemployment Insurance Benefit Cuts on Mental Health: Evidence From Early Pandemic Program Expirations.

Health economics·2026
See all related articles

When estimating population means from cost data, the standard sample mean is a reliable choice. Explicitly modeling lognormal distributions can be less efficient and perform poorly if the true distribution is unknown.

Area of Science:

  • Biostatistics
  • Health Economics
  • Statistical Modeling

Background:

  • Estimating population means from cost data is crucial in health economics and biostatistics.
  • Commentators suggest explicitly modeling cost data distributions for improved efficiency.

Purpose of the Study:

  • To evaluate the efficiency of the sample mean versus a lognormal distribution mean estimator.
  • To compare these estimators using theoretical distributions and empirical datasets.

Main Methods:

  • Simulation experiments were conducted.
  • Evaluated the standard sample mean and a lognormal distribution mean estimator.
  • Utilized theoretical distributions and three large empirical cost datasets.

Main Results:

Related Experiment Videos

  • The sample mean is consistently unbiased.
  • The sample mean is less efficient than the lognormal estimator only when the true distribution is lognormal.
  • The lognormal estimator performed poorly when the true distribution was not lognormal.

Conclusions:

  • In practical scenarios with unknown distributions, the sample mean is generally preferred for estimating population means.
  • Limited sample sizes further support the use of the sample mean over complex distributional modeling.