The iterative process you describe (and again – I’m not sure this happens so consistently – when I buy five apples from the store I don’t buy one, then a second, then a third. I pick up five apples and pay for them all at once), is comparable to numerical optimization methods. So again – if you can get the same answer to an optimization problem using calculus on the one hand or using numerical methods, then FOR MODELING PURPOSES, what is the problem with using calculus?

I’m doing all caps because this is the part of the argument I think you are ignoring. Or do you just think we shouldn’t be modeling?

You’re welcome to think that, but then there’s not much more to say.

]]>