I’m trying to obtain summary stats (average, median,etc) on each quintile (or decile) in a rank chart. For example, median P/E for lowest quintile, median P/E for highest quintile, etc.
What is the most efficient way to do this? I tried isolating the universe based on rank for the factor in question, i.e. frank>0<20, >20<40, >40<60 etc, in combination with aggregate to find the data in the screen output. Unfortunately I ran into some roadblocks.
Can you explain a bit further what your ideal output would be ? You want to see 5 charts, one per bucket, showing two time series each with the performance of the bucket and the median pe at each reconstitution ?
If that’s what you want it’s no easily doable currently.
Sound like an interesting feature , but these things take a long time to do due to the high complexity of designing user interfaces. It might be relatively easy to do using our DataMiner . Have you played with it ? It can run multitudes of operations , collect the data, and produce clean , text output for a particular operation . You can then import into a spreadsheet and generate charts. To do what you want it will require some changes but would be much faster than creating a whole new tool on the website, or shoehorning it into an existing tool.
Marco, I was actually looking for something simpler. I may have confused the issue with talking about rank charts, but what I’m after is for any given date, just the median factor value for each quintile or decile in the universe. i.e. P/E for lowest 10%, P/E for highest 10%.
What you described would also be helpful; I’ll check out the Dataminer, I’ve yet to do this (it’s been on my list !).