![[attach.png]](../idldoc-resources/attach.png)
stats/
grubbs.pro
topgrubbs
STATS
result = grubbs(dat [, CONF=CONF] [, PLOT=PLOT] [, WORDS=WORDS] [, CRIT=CRIT] [, _EXTRA=_EXTRA])
use the Grubbs' test for outlers on an input data set, see http://www.itl.nist.gov/div898/handbook/eda/section3/eda35h.htm
Requires oplot_horiz.pro Requires SSW tree
Return value
1: There are outliers in the data 0: There are no outliers in the data
Parameters
- dat in required
the data to test for outliers
Keywords
- CONF in optional
the confidence level to run the test at
- PLOT in optional
plot the diagnostic plot
- WORDS in optional
display the answer in words
- CRIT in optional
Output the critical cut off value
- _EXTRA in optional
Extra keywords to plot
Examples
IDL> print, grubbs(randomn(123, 1000), conf=0.95, /plot)
1
IDL> print, grubbs(randomn(1027897133, 1000), conf=0.95, /plot)
0
Author information
- History:
Mon Apr 7 20:50:21 2008, Brian Larsen written and tested 2008-SEPT-22, Henry (Trae) Winter Added CRIT keyword.
Statistics
| Lines: | 40 |
| McCabe complexity: |
File attributes
| Modifcation date: | Mon Sep 22 12:37:19 2008 |
| Lines: | 94 |