Scribd
Upload
288 views2 pages

Grubbs' Outlier Test

Grubbs' outlier test checks normally distributed data for outliers. It identifies the value with the largest deviation from the mean and compares it to critical values from tables. If the test statistic is greater than the critical value, the corresponding value is considered an outlier and may be removed. The test can also be conducted one-sided to check if the minimum or maximum value is an outlier.

Uploaded by

David SR
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
288 views2 pages

Grubbs' Outlier Test

Grubbs' outlier test checks normally distributed data for outliers. It identifies the value with the largest deviation from the mean and compares it to critical values from tables. If the test statistic is greater than the critical value, the corresponding value is considered an outlier and may be removed. The test can also be conducted one-sided to check if the minimum or maximum value is an outlier.

Uploaded by

David SR
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 2

Grubbs' Outlier Test

Fundamentals of Statistics contains material of


various lectures and courses of H. Lohninger on
statistics, data analysis and chemometrics......click
here for more.

Home Statistical Tests Outlier Tests Grubbs' Outlier Test Index


See also: Outlier Test - Dean and Dixon, Significance of Outliers 
  ► T Test   ► Chi Square Test   ► Hypothesis Test   ► Anova Test

Grubbs' Outlier Test

Grubbs' outlier test (Grubbs 1969 and Stefansky 1972 ) checks normally distributed data for outliers.
This implies that one has to check whether the data show a normal distribution before applying the Grubbs test.
The Grubbs test always checks the value which shows the largest absolute deviation from the mean. If an
outlier has been identified and removed, the test must not be repeated without adapting the critical value.

The application of the test is quite simple and straightforward: one searches the maximum of the absolute
differences between the values xi and the mean . The result is divided by the standard deviation of the
sample. If the resulting test statistic g is greater than the critical value, the corresponding value can be regarded
to be an outlier. An extract of the critical values is shown in the following table:

gcrit gcrit gcrit gcrit gcrit gcrit


n   n   n
a=0.05 a=0.01 a=0.05 a=0.01 a=0.05 a=0.01
3 1.1543 1.1547   15 2.5483 2.8061   80 3.3061 3.6729
4 1.4812 1.4962   16 2.5857 2.8521   90 3.3477 3.7163
5 1.7150 1.7637   17 2.6200 2.8940   100 3.3841 3.7540
6 1.8871 1.9728   18 2.6516 2.9325   120 3.4451 3.8167
7 2.0200 2.1391   19 2.6809 2.9680   140 3.4951 3.8673
8 2.1266 2.2744   20 2.7082 3.0008   160 3.5373 3.9097
9 2.2150 2.3868   25 2.8217 3.1353   180 3.5736 3.9460
10 2.2900 2.4821   30 2.9085 3.2361   200 3.6055 3.9777
11 2.3547 2.5641   40 3.0361 3.3807   300 3.7236 4.0935
12 2.4116 2.6357   50 3.1282 3.4825   400 3.8032 4.1707
13 2.4620 2.6990   60 3.1997 3.5599   500 3.8631 4.2283
14 2.5073 2.7554   70 3.2576 3.6217   600 3.9109 4.2740

http://www.statistics4u.com/fundstat_eng/ee_grubbs_outliertest.html[15/09/2015 11:18:23]
Grubbs' Outlier Test

There is a one-sided alternative which allows to test either the minimum xmin or the maximum xmax of the entire
data set. The test statistics calculates according to the following formulas:

A value can be regarded an outlier if the statistic g is greater than the critical value. Please note that in the case
of the one-sided test the critical values are different. An extract is given below:

gcrit gcrit gcrit gcrit gcrit gcrit


n   n   n
a=0.05 a=0.01 a=0.05 a=0.01 a=0.05 a=0.01
3 1.1531 1.1546   15 2.4090 2.7049   80 3.1319 3.5208
4 1.4625 1.4925   16 2.4433 2.7470   90 3.1733 3.5632
5 1.6714 1.7489   17 2.4748 2.7854   100 3.2095 3.6002
6 1.8221 1.9442   18 2.5040 2.8208   120 3.2706 3.6619
7 1.9381 2.0973   19 2.5312 2.8535   140 3.3208 3.7121
8 2.0317 2.2208   20 2.5566 2.8838   160 3.3633 3.7542
9 2.1096 2.3231   25 2.6629 3.0086   180 3.4001 3.7904
10 2.1761 2.4097   30 2.7451 3.1029   200 3.4324 3.8220
11 2.2339 2.4843   40 2.8675 3.2395   300 3.5525 3.9385
12 2.2850 2.5494   50 2.9570 3.3366   400 3.6339 4.0166
13 2.3305 2.6070   60 3.0269 3.4111   500 3.6952 4.0749
14 2.3717 2.6585   70 3.0839 3.4710   600 3.7442 4.1214

Who Controls The World?


One Group Rules In Secret This New Report Exposes All

Home Statistical Tests Outlier Tests Grubbs' Outlier Test

Last Update: 2012-10-08

http://www.statistics4u.com/fundstat_eng/ee_grubbs_outliertest.html[15/09/2015 11:18:23]

You might also like