Thank you Joachim.
I don't think your method contradicts using forward ratios as a time series for one stock. Comparing current forward ratio to its historical average and assuming mean reversion seems to me as a valid check for under/overpricing. What are your thoughts?