I am looking at the days in the month a person gets paid over a period of time. I want to determine if the variance amongst days is much or not. If its much, I want to flag that person. For example, lets say a person gets paid on these days in 5 months,
[28, 29, 30, 28, 10]
I want to flag this person because of the day 10. 10 is quite distant from other days. If I calculate the variance, I get 56.8 but I am confused as to where to define a threshold that says something is wrong.
a = np.array([28, 29, 30, 28, 10])
np.var(a)
Or is there a better way of achieving this?
