Trimmed mean for grouped data

Originally Posted by Duarte_RV

I have a dataset that looks like this but with 800,00 rows
[....]
I want to calculate the average time per each postcode but, as there are some outliers, I need the trimmed mean. I know how to use the =TRIMMEAN formula, but I don't know how to group per postcode and then calculate the trimmed mean. [....]

I would also need to add some more levels of complexity, like calculating the trimmed mean for a specific postcode in a specific month.

First, your table is not helpful because it does not include actual Excel row and column labels (1, 2 etc; and A, B etc). It is better to attach an example Excel file.

Second, TRIMMEAN does not remove outliers per se. Instead, it arbitrarily removes "p" percent of the data (the first and last p/2 percent), where "p" is the percentage specified in the second parameter. Also beware of how TRIMMEAN rounds the number of data to exclude. See the TRIMMEAN help page for details.

If you want to know about outliers, I could explain further.

But if you insist on using TRIMMEAN, conceptually you might array-enter something like this (press ctrl+shift+Enter instead of just Enter):

=TRIMMEAN(IF(D2:D80001="CB1",IF(B2:B80001=1,F2:F80001)), 1%)

to remove 800 data points (80000*1%), the first and last 400 data points, from the January data (month 1) with postal code "CB1".