Grubbs's test: Difference between revisions

Content deleted Content added

Inline

Revision as of 15:35, 18 January 2024

In statistics, Grubbs's test or the Grubbs test (named after Frank E. Grubbs, who published the test in 1950^[1]), also known as the maximum normalized residual test or extreme studentized deviate test, is a test used to detect outliers in a univariate data set assumed to come from a normally distributed population.

Definition

Grubbs's test is based on the assumption of normality. That is, one should first verify that the data can be reasonably approximated by a normal distribution before applying the Grubbs test.^[2]

Grubbs's test detects one outlier at a time. This outlier is expunged from the dataset and the test is iterated until no outliers are detected. However, multiple iterations change the probabilities of detection, and the test should not be used for sample sizes of six or fewer since it frequently tags most of the points as outliers.^[3]

Grubbs's test is defined for the following hypotheses:

H₀: There are no outliers in the data set

H_a: There is exactly one outlier in the data set

The Grubbs test statistic is defined as

G={\frac {\displaystyle \max _{i=1,\ldots ,N}\left\vert Y_{i}-{\bar {Y}}\right\vert }{s}}

with ${\overline {Y}}$ and $s$ denoting the sample mean and standard deviation, respectively. The Grubbs test statistic is the largest absolute deviation from the sample mean in units of the sample standard deviation.

This is the two-sided test, for which the hypothesis of no outliers is rejected at significance level α if

G>{\frac {N-1}{\sqrt {N}}}{\sqrt {\frac {t_{\alpha /(2N),N-2}^{2}}{N-2+t_{\alpha /(2N),N-2}^{2}}}}

with t_{α/(2N),N−2} denoting the upper critical value of the t-distribution with N − 2 degrees of freedom and a significance level of α/(2N).

One-sided case

Grubbs's test can also be defined as a one-sided test, replacing α/(2N) with α/N. To test whether the minimum value is an outlier, the test statistic is

G={\frac {{\bar {Y}}-Y_{\min }}{s}}

with Y_min denoting the minimum value. To test whether the maximum value is an outlier, the test statistic is

G={\frac {Y_{\max }-{\bar {Y}}}{s}}

with Y_max denoting the maximum value.

Related techniques

Several graphical techniques can be used to detect outliers. A simple run sequence plot, a box plot, or a histogram should show any obviously outlying points. A normal probability plot may also be useful.

References

^ Grubbs, Frank E. (1950). "Sample criteria for testing outlying observations". Annals of Mathematical Statistics. 21 (1): 27–58. doi:10.1214/aoms/1177729885. hdl:2027.42/182780.
^ Quoted from the Engineering and Statistics Handbook, paragraph 1.3.5.17, http://www.itl.nist.gov/div898/handbook/eda/section3/eda35h.htm
^ Adikaram, K. K. L. B.; Hussein, M. A.; Effenberger, M.; Becker, T. (2015-01-14). "Data Transformation Technique to Improve the Outlier Detection Power of Grubbs's Test for Data Expected to Follow Linear Relation". Journal of Applied Mathematics. 2015: 1–9. doi:10.1155/2015/708948.

@@ Line 1: / Line 1: @@
-In statistics, '''Grubbs' test'''<!--possessive of the proper noun--> or the '''Grubbs test'''<!--attributive use of the proper name--> (named after [[Frank E. Grubbs]], who published the test in 1950<ref>{{cite journal |last=Grubbs |first=Frank E. |title=Sample criteria for testing outlying observations |journal=[[Annals of Mathematical Statistics]] |volume=21 |issue=1 |pages=27–58 |doi=10.1214/aoms/1177729885 |year=1950 |doi-access=free |hdl=2027.42/182780 |hdl-access=free }}</ref>), also known as the '''maximum normalized [[errors and residuals in statistics|residual]] test''' or '''extreme studentized deviate test''', is a [[Statistical hypothesis testing|test]] used to detect [[outlier]]s in a [[univariate]] data set assumed to come from a [[normal distribution|normally distributed]] population.
+In statistics, '''Grubbs's test'''<!--possessive of the proper noun--> or the '''Grubbs test'''<!--attributive use of the proper name--> (named after [[Frank E. Grubbs]], who published the test in 1950<ref>{{cite journal |last=Grubbs |first=Frank E. |title=Sample criteria for testing outlying observations |journal=[[Annals of Mathematical Statistics]] |volume=21 |issue=1 |pages=27–58 |doi=10.1214/aoms/1177729885 |year=1950 |doi-access=free |hdl=2027.42/182780 |hdl-access=free }}</ref>), also known as the '''maximum normalized [[errors and residuals in statistics|residual]] test''' or '''extreme studentized deviate test''', is a [[Statistical hypothesis testing|test]] used to detect [[outlier]]s in a [[univariate]] data set assumed to come from a [[normal distribution|normally distributed]] population.
 ==Definition==
-Grubbs' test is based on the assumption of [[normal distribution|normality]]. That is, one should first verify that the data can be reasonably approximated by a normal distribution before applying the Grubbs test.<ref>Quoted from the ''Engineering and Statistics Handbook'', paragraph 1.3.5.17, http://www.itl.nist.gov/div898/handbook/eda/section3/eda35h.htm</ref>
+Grubbs's test is based on the assumption of [[normal distribution|normality]]. That is, one should first verify that the data can be reasonably approximated by a normal distribution before applying the Grubbs test.<ref>Quoted from the ''Engineering and Statistics Handbook'', paragraph 1.3.5.17, http://www.itl.nist.gov/div898/handbook/eda/section3/eda35h.htm</ref>
-Grubbs' test detects one outlier at a time. This outlier is expunged from the dataset and the test is iterated until no outliers are detected. However, multiple iterations change the probabilities of detection, and the test should not be used for sample sizes of six or fewer since it frequently tags most of the points as outliers.<ref>{{Cite journal|last1=Adikaram|first1=K. K. L. B.|last2=Hussein|first2=M. A.|last3=Effenberger|first3=M.|last4=Becker|first4=T.|date=2015-01-14|title=Data Transformation Technique to Improve the Outlier Detection Power of Grubbs' Test for Data Expected to Follow Linear Relation|journal=Journal of Applied Mathematics|volume=2015|pages=1–9|language=en|doi=10.1155/2015/708948|doi-access=free}}</ref>
+Grubbs's test detects one outlier at a time. This outlier is expunged from the dataset and the test is iterated until no outliers are detected. However, multiple iterations change the probabilities of detection, and the test should not be used for sample sizes of six or fewer since it frequently tags most of the points as outliers.<ref>{{Cite journal|last1=Adikaram|first1=K. K. L. B.|last2=Hussein|first2=M. A.|last3=Effenberger|first3=M.|last4=Becker|first4=T.|date=2015-01-14|title=Data Transformation Technique to Improve the Outlier Detection Power of Grubbs's Test for Data Expected to Follow Linear Relation|journal=Journal of Applied Mathematics|volume=2015|pages=1–9|language=en|doi=10.1155/2015/708948|doi-access=free}}</ref>
-Grubbs' test is defined for the following [[statistical hypothesis|hypotheses]]:
+Grubbs's test is defined for the following [[statistical hypothesis|hypotheses]]:
 :H<sub>0</sub>: There are no outliers in the data set
 :H<sub>a</sub>: There is exactly one outlier in the data set
@@ Line 25: / Line 25: @@
 ===One-sided case===
-Grubbs' test can also be defined as a one-sided test, replacing α/(2''N'') with α/''N''. To test whether the minimum value is an outlier, the test statistic is
+Grubbs's test can also be defined as a one-sided test, replacing α/(2''N'') with α/''N''. To test whether the minimum value is an outlier, the test statistic is
 :<math>
 G = \frac{\bar{Y}-Y_\min}{s}