there was a question on which method is best if normal dist assumptions are valid. I chose analytical method but the choice had monte carlo as an option as well. Monte carlo can operate under any assumption. Is the question not correctly formed or am I missing something?
Yeah but if you know that the distribution is normal you can do calculations that are statistically optimal. You can do Monte Carlo too, but you can run your computer for 6 months and only get results that are nearly as good as the analytical results.
if problem can be solved analytically, why would anyone want to solve it numerically?
yeah I understand that it is easy in terms of calculations, less comp intensive etc but the inclusion of monte carlo , which works under all kinds of dist assumptions in the answer choices made be wonder if the question itself has been correctly formed… let me put it this way. what if I chose monte carlo. does it make me incorrect ?
Because just because it can be done analytically doesn’t mean that anyone wants to think hard enough to do it that way. I don’t solve differential equations anymore, I don’t compute definite integrals, I don’t analytically find standard errors, etc. even if I can because my computer can do them numerically faster and close enough for all reasonable uses without my thinking much at all.
krishna1 Wrote: ------------------------------------------------------- > yeah I understand that it is easy in terms of > calculations, less comp intensive etc but the > inclusion of monte carlo , which works under all > kinds of dist assumptions in the answer choices > made be wonder if the question itself has been > correctly formed… > > let me put it this way. what if I chose monte > carlo. does it make me incorrect ? On the test, probably. In the real world, no.