0
$\begingroup$

Imagine this example. I've a field with following values:

10000
13455
90
12347
90
1111
235679

How can I identify that 90 are a outlier? Using average or standard deviation?

Thanks!

$\endgroup$
  • 2
    $\begingroup$ What makes it an outlier for your purposes? $\endgroup$ – Glen_b Aug 26 '16 at 1:01
  • $\begingroup$ You usually need a mean to calculate a standard deviation. $\endgroup$ – david25272 Aug 26 '16 at 3:22
1
$\begingroup$

We can plot an outlier box plot with fences to graphically display our data. Then we can define quantities for identifying extreme values in the tails of the distribution. For instance, if Q1 is the first quartile, Q3 is the third quartile and the interquartile range is IQ = Q3-Q1 then

lower inner fence: Q1 - 1.5*IQ
upper inner fence: Q3 + 1.5*IQ
lower outer fence: Q1 - 3*IQ
upper outer fence: Q3 + 3*IQ

Finally, we can apply an outlier detection criteria. A point beyond an inner fence on either side is considered a mild outlier. A point beyond an outer fence is considered an extreme outlier.

See example of an outlier box plot.

| cite | improve this answer | |
$\endgroup$

Not the answer you're looking for? Browse other questions tagged or ask your own question.