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

Multivariate adaptive regression splines: a powerful method for detecting disease-risk relationship differences among

Timothy P York1, Lindon J Eaves, Edwin J C G van den Oord

  • 1Massey Cancer Center, Virginia Commonwealth University, Virginia Institute for Psychiatric and Behavioral Genetics, P.O. Box 980003, Richmond, VA 23298-0003, USA. tpyork@vcu.edu

Statistics in Medicine
|August 16, 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

Predicting Functional Changes in Down Syndrome During the COVID-19 Pandemic: The Role of Biopsychosocial Determinants of Health.

medRxiv : the preprint server for health sciences·2026
Same author

Caregiving Demands and Depression Symptoms among Caregivers of Individuals with Down Syndrome during the COVID-19 Pandemic.

medRxiv : the preprint server for health sciences·2026
Same author

Methylome-wide association study in blood suggests cell type-specific relationships between DNA methylation and lifetime anxiety.

Nature communications·2026
Same author

A large-scale DNA methylation study of alcohol use identified robust associations and cell-type specific insights.

Molecular psychiatry·2026
Same author

Risk factors for spontaneous preterm birth are mediated through changes in cervical length.

Communications medicine·2025
Same author

Cell type-specific methylome-wide association studies of childhood ADHD symptoms.

European neuropsychopharmacology : the journal of the European College of Neuropsychopharmacology·2025
Same journal

Latent Class Log-Linear Models for Estimating Diagnostic Test Accuracy Without a Gold Standard: A Simulation Study.

Statistics in medicine·2026
Same journal

Interpretable Bayesian Modeling for Multireader Multicase Studies: Addressing Overdispersion and Limited Sample Size in Diagnostic Enhancement Evaluation.

Statistics in medicine·2026
Same journal

Adaptive Sequential Multiple Hypotheses Testing for Concomitant Vaccine Safety Surveillance.

Statistics in medicine·2026
Same journal

Novel Distance Regression for Repeated Outcomes With Missing Data: Applications to Longitudinal and Crossover Studies of Microbiome Beta-Diversity.

Statistics in medicine·2026
Same journal

Optimal Weighted Tests for Replication Studies and the 'Two-Trials Rule' With Multiple Hypotheses.

Statistics in medicine·2026
Same journal

Identifiable Copula-Double-Cox Models: A Fully Parametric Framework for Dependent Right-Censored Survival Data.

Statistics in medicine·2026
See all related articles

Multivariate adaptive regression splines (MARS) offer superior power for detecting differences in regression curves between subgroups compared to polynomial methods. This advantage holds across various curve complexities, especially for non-linear relationships in medical research.

Area of Science:

  • Medical research
  • Biostatistics
  • Statistical modeling

Background:

  • Analyzing subgroup differences in regression curves is crucial in medical research, such as for drug treatment effects or genotype-environment interactions.
  • Exploratory techniques are often necessary due to a lack of prior knowledge about regression curve shapes and their variations across subgroups.

Purpose of the Study:

  • To compare the power of multivariate adaptive regression splines (MARS) and polynomial least squares curve fitting for detecting differences in regression curves between subgroups.
  • To evaluate these methods across linear, logistic, and complex non-linear simulated data.

Main Methods:

  • Simulations were conducted using linear, logistic, and complex non-linear regression curves.
  • The statistical power of MARS and polynomial methods was assessed and compared.

Related Experiment Videos

Main Results:

  • MARS demonstrated, on average, 1.4 times higher statistical power than polynomial methods.
  • MARS showed increased power even for linear regression curves.
  • The performance gains for MARS increased with the complexity of the regression curve.

Conclusions:

  • MARS is a more powerful exploratory technique than polynomial fitting for identifying differences in regression curves across subgroups in medical research.
  • For highly non-linear curves, model-free methods like MARS may be the only viable option.