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

A scan statistic with a variable window

N Nagarwalla1

  • 1Department of Oral Pathology, Harvard School of Dental Medicine and Center for Blood Research, Harvard Medical School, Boston, MA 02115, USA.

Statistics in Medicine
|April 15, 1996
PubMed
Summary

This study introduces a flexible scan statistic using a variable window size for detecting event clusters. This method improves upon fixed-width statistics by allowing for variable cluster sizes, enhancing epidemiological analysis.

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

Use of recombinant pemphigus vulgaris antigen in development of ELISA and IB assays to detect pemphigus vulgaris autoantibodies.

Journal of the European Academy of Dermatology and Venereology : JEADV·1998
Same author

Oral pemphigoid. Subset of cicatricial pemphigoid?

Oral surgery, oral medicine, oral pathology, oral radiology, and endodontics·1998
Same author

Childhood leukaemia in Sweden: using GIS and a spatial scan statistic for cluster detection.

Statistics in medicine·1996
Same author

Correlation of peptide specificity and IgG subclass with pathogenic and nonpathogenic autoantibodies in pemphigus vulgaris: a model for autoimmunity.

Proceedings of the National Academy of Sciences of the United States of America·1995
Same author

Spatial disease clusters: detection and inference.

Statistics in medicine·1995
Same author

Antibodies to the 70-kDa heat-shock protein in patients with thermal burns.

Clinical immunology and immunopathology·1995

Area of Science:

  • Statistics
  • Epidemiology
  • Biostatistics

Background:

  • The simple scan statistic detects clusters of events within a fixed-width interval.
  • Fixed-width selection can limit the detection of clusters with varying sizes.
  • This poses challenges in epidemiological studies where cluster sizes may not be predetermined.

Purpose of the Study:

  • To propose and evaluate a novel scan statistic employing a variable window.
  • To address the limitations of fixed-width scan statistics in cluster detection.
  • To provide a more adaptable statistical tool for identifying non-random event clustering.

Main Methods:

  • Development of a generalized likelihood ratio test for cluster detection.
  • Implementation of a scan statistic with a variable window size.
  • Application of the method to analyze birth defects data.

Main Results:

  • The proposed variable-width scan statistic effectively identifies clusters without prior width specification.
  • The method offers greater flexibility compared to traditional fixed-width approaches.
  • Analysis of birth defects data demonstrated the utility of the variable window scan statistic.

Conclusions:

  • A variable window scan statistic provides a powerful and flexible alternative for detecting event clusters.
  • This approach enhances the ability to identify non-random clustering in epidemiological and other fields.
  • The generalized likelihood ratio test offers a robust framework for cluster analysis with adaptable window sizes.

Related Experiment Videos