get_critical_value#

pycafee.sample.outliers.Dixon.get_critical_value(self, n_rep, ratio=None, alfa=None)#

This function returns the critical value (tabulated) of the Dixon test for outlier detection [1]. The critical values are the values given by Rorabacher [2] for the two-tailed distribution.

Parameters

n_repint: The total number of observations (n_rep >= 3, vary).
ratiostr, optional: The ratio. It can be "r10" (or None), "r11", "r12", "r20", "r21" or "r22".
alfafloat: The significance level (between 0.0 and 1.0, default = 0.05)

Returns

resulttuple with:

criticalfloat: The critical value.
alfafloat: The corresponding significance level.

Notes

There are critical values for confidence levels of 0.20, 0.10, 0.05, 0.04, 0.02 and 0.01, which also depend on the ratio parameter and the sample size (n_rep):

For ratio="r10" there are critical values within this range 3<=n_rep<=30;
For ratio="r11" there are critical values within this range 4<=n_rep<=30;
For ratio="r12" there are critical values within this range 5<=n_rep<=30;
For ratio="r20" there are critical values within this range 4<=n_rep<=30;
For ratio="r21" there are critical values within this range 5<=n_rep<=30;
For ratio="r22" there are critical values within this range 6<=n_rep<=30;

References

1: DIXON, W. J. Processing Data for Outliers. Biometrics, v. 9, n. 1, p. 74–89, 1953.
2: RORABACHER, D. B. Statistical Treatment for Rejection of Deviant Values: Critical Values of Dixon’s “Q” Parameter and Related Subrange Ratios at the 95% Confidence Level. v. 63, n. 2, p. 139–146, 1991.

Examples

>>> from pycafee.sample.dixon import Dixon
>>> outliers = Dixon()
>>> result = outliers.get_critical_value(7)
>>> print(result)
DixonResult(critical=0.568, alpha=0.05)

>>> from pycafee.sample.dixon import Dixon
>>> outliers = Dixon()
>>> result = outliers.get_critical_value(7, alfa=0.01)
>>> print(result)
DixonResult(critical=0.51, alpha=0.01)

>>> from pycafee.sample.dixon import Dixon
>>> outliers = Dixon()
>>> result = outliers.get_critical_value(7, ratio="r11")
>>> print(result)
DixonResult(critical=0.673, alpha=0.05)

fit

_r10