ColumnRank by group?

[#1]

Hello,

Is there a way to build a formula equivalent to «ColumnRank by…» (group). I normally use the distribution platform (with «by group» option) and save ranks for that purpose in a static column. But this time, I need a dynamic column (calculated).

Thanks,

Claude

[#2]

Thanks Ron.

Finally, I chose to modify the original update script. It's a 3 steps procedure:

1. The first step is the original update (ODBC data source) which is creating a new table (because of the hard code columns);
2. Then a second script step is updating my main table with this new table's data;
3. The last step is closing the new table.

This way I can to create as many new hard code columns as I need (the script won't need any fix).

Claude

[#3 · Accepted solution]

Hi all,

The Col Rank function does now support BY variables as was originally requested. It looks like full support was added in the JMP 13 timeframe, so if you're using that version or a more recent one, you should be able to do this.

The online JMP Help gives a few more details and tie-break options:

https://www.jmp.com/support/help/en/15.1/#page/jmp/statistical-functions-2.shtml?os=win&source=appli...

I know this is an older thread, but wanted this information available for anyone else who may be looking for it.

Cheers!

Audrey