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

Partial knowledge, entropy, and estimation.

J Macqueen1, J Marschak

  • 1Western Management Science Institute, University of California, Los Angeles, Los Angeles, Calif. 90024.

Proceedings of the National Academy of Sciences of the United States of America
|October 1, 1975
PubMed
Summary

This study evaluates the maximum entropy method for estimating probability distributions using partial knowledge. It finds that while one justification is rejected, another, related to statistical thermodynamics, is valid under specific conditions.

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

Syphilis Insontium in Palestine.

The British journal of venereal diseases·2011
Same author

Sounding board: in support of part-time faculty.

Nursing outlook·1981
Same author

Colorimetric microdetermination of plasma lactate.

The American journal of medical technology·1979
Same author

A review of clinical applications and methods for cholinesterase.

The American journal of medical technology·1973
Same author

A simplified colorimetric method for serum isocitrate dehydrogenase.

The American journal of medical technology·1972
Same author

To the editor.

The American journal of medical technology·1972

Area of Science:

  • Information Theory
  • Statistical Modeling
  • Probability Distributions

Background:

  • The maximum entropy principle is widely used to estimate probability distributions when only partial information is available.
  • This method has been applied across diverse fields, including economics, transportation, finance, and marketing.
  • Two common justifications for employing the maximum entropy method are its 'conservatism' and its connection to statistical thermodynamics.

Purpose of the Study:

  • To critically examine and respond to two primary justifications for using the maximum entropy method.
  • To determine the validity of the 'conservatism' argument and the statistical thermodynamics argument for maximum entropy estimation.

Main Methods:

  • The study analyzes the theoretical underpinnings of the maximum entropy principle.

Related Experiment Videos

  • It evaluates the mathematical and conceptual justifications provided in existing literature.
  • The analysis focuses on the conditions under which the statistical thermodynamics analogy holds true.

    • Main Results:

    • The justification based on maximizing 'uncertainty' as a conservative approach is rejected.
    • The justification linking maximum entropy to the mathematics of statistical thermodynamics is deemed valid.
    • This validity is contingent upon specific definitions of 'complete ignorance' and particular forms of constraints and loss functions.

    Conclusions:

    • The maximum entropy method's theoretical foundation is strengthened by its connection to statistical thermodynamics under defined conditions.
    • The 'conservatism' argument for maximum entropy is not a sufficient justification on its own.
    • Further research may explore the implications of these findings for practical applications in various scientific domains.