Scan test
|
|
|
|
Indications/Recommendations for use: When cases are clustered the maximum number of cases appearing in the scanning window
will be large (more).
|
|
Description: A test for time clustering in a single time series.
|
Test statistic: The test statistic, Sw, is the maximum number of cases appearing in a window as it is
moved continuously along a time series. For example, for this time series
0 1 2 0 0 2 0 1
^^^
and a window width of 2, Sw is 3. The significance of Sw is obtained by numerical approximation.
|
Sw: the maximum number of cases appearing in a window as
it is moved continuously along a time series.
|
|
Null Hypothesis:
Cases occur at random through time.
|
|
|
Alternative Hypothesis:
Cases cluster some time periods.
|
|
GeoMed Inputs:
A time series of case counts and the scanning window width.
|
|
GeoMed Outputs:
|
|
Example Analysis
|
Reference:
Wallenstein, S. 1980. A test for detection of clustering over time. American Journal of Epidemiology, 104: 576-584.
Wallenstein, S. and N. Neff. 1987. An approximation for the distribution of the scan statistic. Statistics in Medicine, 6: 197-207.
|
