Determine Outlier with increasing mean

[Senor Edward]

What is the best methodology to determine whether a data point is an outlier of a data set when the data has an increasing mean?

Current approach is to calculate the difference in standard deviations between the current point and the population's mean before this point. If it is more than 4 standard deviations away, I determine this to be an outlier. However, this methodology won't be the most effective in the scenario where the mean of the population is gradually increasing over time.

Thoughts?

[Jay Westreich]

For boxplot purposes, Stata determines an outlier to be any point that is greater than 1.5 times the interquartile range away from the interquartile range. You can create a variable with each data point's percentile using egen pctile and create a variable with the interquartile range using egen iqr. Both of these commands allow bysort.