Dixon tests for outlier
Description:
Performs several variants of Dixon test for detecting outlier in
data sample.
Usage:
[pval,Q] = dixontest(x,type,opposite,twosided)
Arguments:
x: a numeric vector or matrix of data values. Each column of a
matrix is treated as independent sample set.
opposite: a logical (0,1) indicating whether you want to check not the value
with largest difference from the mean, but opposite (lowest,
if most suspicious is highest etc.). Default 0.
type: an integer specyfying the variant of test to be performed.
Possible values are compliant with these given by Dixon
(1950): 10, 11, 12, 20, 21. If this value is set to zero, a
variant of the test is chosen according to sample size (10
for 3-7, 11 for 8-10, 21 for 11-13, 22 for 14 and more). The
lowest or highest value is selected automatically, and can be
reversed used 'opposite' parameter.
two.sided: treat test as two-sided (default=1).
Details:
The p-value is calculating by interpolation using 'dixoncdf' and
'qtable'. According to Dixon (1951) conclusions, the critical
values can be obtained numerically only for n=3. Other critical
values are obtained by simulations, taken from original Dixon's
paper, and regarding corrections given by Rorabacher (1991).
Value:
Q: the value of Dixon Q-statistic.
pval: the p-value for the test.
Author(s):
Lukasz Komsta, ported from R package "outliers".
See R News, 6(2):10-13, May 2006
References:
Dixon, W.J. (1950). Analysis of extreme values. Ann. Math. Stat.
21, 4, 488-506.
Dixon, W.J. (1951). Ratios involving extreme values. Ann. Math.
Stat. 22, 1, 68-78.
Rorabacher, D.B. (1991). Statistical Treatment for Rejection of
Deviant Values: Critical Values of Dixon Q Parameter and Related
Subrange Ratios at the 95 percent Confidence Level. Anal. Chem.
83, 2, 139-146.
