I agree that things are often ephemeral to some extent. I'd say modelling trading impact and trying to optimise around it is actually quite intense modelling o

I agree that things are often ephemeral to some extent. I'd say modelling trading impact and trying to optimise around it is actually quite intense modelling of supply and demand at times. But have you actually tried to establish elasticity and run experiments to that end? In all my time in industry, I have mostly not seen efforts to do that, but when people did that it actually was quite successful in increasing revenue. Fixed the link - thank you.