Search research articles

ABOUT JoVE

Overview Leadership Blog JoVE Help Center

AUTHORS

Publishing Process Editorial Board Scope & Policies Peer Review FAQ Submit

LIBRARIANS

Testimonials Subscriptions Access Resources Library Advisory Board FAQ

RESEARCH

JoVE Journal Methods Collections JoVE Encyclopedia of Experiments Archive

EDUCATION

JoVE Core JoVE Business JoVE Science Education JoVE Lab Manual Faculty Resource Center Faculty Site

Terms & Conditions of Use

Search research articles

Related Experiment Videos

An uncertainty measure in logistic discrimination.

E Lesaffre¹, A Albert

¹Department of Medical Informatics, U.Z. Gasthuisberg, University of Leuven, Belgium.

Statistics in Medicine

|April 1, 1988

Summary

Sampling variation impacts logistic discriminant analysis error rates. A new beta-confidence allocation rule assesses classification accuracy and stability, enhancing posterior probability interpretation for medical diagnoses.

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

Gemcitabine and low dose carboplatin in the treatment of elderly patients with advanced non-small cell lung cancer.

Lung cancer (Amsterdam, Netherlands)·2003

Same author

Structure of bacteriocin AS-48: from soluble state to membrane bound state.

Journal of molecular biology·2003

Same author

Crural artery bypass with the autogenous greater saphenous vein.

European journal of vascular and endovascular surgery : the official journal of the European Society for Vascular Surgery·2003

Same author

[Predictive value of prenatal MRI in the diagnosis of thoracic congenital malformations].

Cirugia pediatrica : organo oficial de la Sociedad Espanola de Cirugia Pediatrica·2003

Same author

Weekly paclitaxel for advanced non-small cell lung cancer patients not suitable for platinum-based therapy.

Neoplasma·2003

Same author

Factors predictive of nephropathy in DCCT Type 1 diabetic patients with good or poor metabolic control.

Diabetic medicine : a journal of the British Diabetic Association·2003

Area of Science:

Statistics
Biostatistics
Medical Decision Making

Background:

Logistic discriminant analysis is susceptible to sampling variation, affecting classification accuracy.
Assessing the reliability of individual classifications is crucial for decision-making in clinical settings.
Existing methods may not adequately quantify the uncertainty associated with classification rules.

Purpose of the Study:

To investigate the impact of sampling variation on logistic discriminant analysis.
To introduce and evaluate the beta-confidence allocation rule for assessing classification accuracy.
To propose a method for associating credibility measures with posterior probabilities.

Main Methods:

Discussion of sampling variation effects on logistic discriminant analysis.
Introduction of the beta-confidence allocation rule for significance testing of assignments.
Application of the procedure to sample data to evaluate classification rule stability.

Main Results:

The beta-confidence allocation rule enables testing of correct/incorrect observations at a specified significance level.
The procedure provides insights into the sharpness and stability of the estimated classification rule.
Individual posterior probabilities can be assigned a credibility measure.

Conclusions:

The beta-confidence allocation rule offers a robust method for evaluating classification accuracy in logistic discriminant analysis.
This approach enhances the reliability of diagnostic or predictive models by quantifying uncertainty.
The proposed method has practical applications, as demonstrated in liver disease patient classification.

Related Experiment Videos