Chi-squared test for outlier
Description:
Performs a chisquared test for detection of one outlier in a
vector.
Usage:
[pval,chisq] = chisqouttest(x,variance,opposite)
Arguments:
x: a numeric vector of data values.
variance: known variance of population. if not given, estimator from
sample is taken, but there is not so much sense in such test
(it is similar to z-scores)
opposite: a logical indicating whether you want to check not the value
with largest difference from the mean, but opposite (lowest,
if most suspicious is highest etc.)
Details:
This function performs a simple test for one outlier, based on
chisquared distribution of squared differences between data and
sample mean. It assumes known variance of population. It is
rather not recommended today for routine use, because several more
powerful tests are implemented (see other functions mentioned
below). It was discussed by Dixon (1950) for the first time, as
one of the tests taken into account by him.
Value:
chisq: the value of chisquared-statistic.
pval: the p-value for the test.
Note:
This test is known to reject only extreme outliers, if no known
variance is specified.
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.
