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 regression trees for analysis of abundance data.

David R Larsen1, Paul L Speckman

  • 1Department of Forestry, University of Missouri, Columbia, Missouri 65211, USA. LarsenDR@missouri.edu

Biometrics
|June 8, 2004
PubMed
Summary

This study introduces multivariate regression trees to predict co-occurring plant species abundance in forests. This method offers a new way to analyze complex ecological data and understand species interactions.

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

Gradual growth versus shape invariance in perceptual decision making.

Psychological review·2010
Same author

Bayesian t tests for accepting and rejecting the null hypothesis.

Psychonomic bulletin & review·2009
Same author

A hierarchical process-dissociation model.

Journal of experimental psychology. General·2008
Same author

Detecting chance: a solution to the null sensitivity problem in subliminal priming.

Psychonomic bulletin & review·2007
Same author

Hierarchical models for the probabilities of conception.

Biometrical journal. Biometrische Zeitschrift·2006
Same author

An evaluation of the Vincentizing method of forming group-level response time distributions.

Psychonomic bulletin & review·2004

Area of Science:

  • Ecology
  • Forestry
  • Statistical Modeling

Background:

  • Predicting the abundance of co-occurring plant species is crucial for understanding forest ecosystems.
  • Existing methods may not adequately capture the simultaneous relationships between multiple dependent variables.

Purpose of the Study:

  • To develop and illustrate a novel multivariate regression tree methodology.
  • To predict the abundance of several co-occurring plant species in Missouri Ozark forests.
  • To offer an alternative to traditional cluster analysis for specific research questions.

Main Methods:

  • The study employs a variation of multivariate regression tree methodology, adapted from Segal (1992).
  • This technique models the simultaneous co-occurrence of multiple dependent variables.

Related Experiment Videos

  • It utilizes independent variables to define clusters that are homogeneous with respect to dependent variables.

    • Main Results:

    • The multivariate regression tree methodology was successfully applied to predict plant species abundance in a forest setting.
    • The technique demonstrated its capability in handling multiple co-occurring dependent variables.
    • Results highlight the potential for this method in ecological and other multivariate analyses.

    Conclusions:

    • Multivariate regression trees provide a powerful tool for predicting simultaneous co-occurrences of dependent variables.
    • This methodology offers a flexible alternative to cluster analysis when independent variables define homogeneous groups.
    • The approach has broad applicability across various scientific disciplines requiring multivariate data analysis.