S11.1 Simulation

[@faresalouf]

amazing explanation thank you

[@abir95571]

inverse of a CDF is supposed to take probability and give the respective R.V as an output ... why's the prof referring u as a RV when it should be some prob. I don't get it

[@paulgoyes3335]

If I generated 100 samples per each variable X, Y, Z, so... The simulation must be just the combination of each x, y, z? That's mean, g(X, Y, Z) = G will be a set of 100*100*100 values, right?

[@k10shetty]

I believe for one value of U(0,1), evaluate X, Y, and Z, then calculate Z, and go on

[@まつまつ-x1f]

Hmm...this is quite a calculus...

[@AbhyudayaRanglani]

Most lectures upto this point were well presented, but this one is all over the place. I have no idea of what was being said/implied and/or why such an important topic was treated so poorly.

[@paulgoyes3335]

I understood that... If you have a very complicated process given by a function g(X, Y, Z), but you know that those variables have a specific distribution... So you can guess the distribution of g(X, Y, Z) by a simulation. If g is not so complicate, you can do of course the steps described before. Is my opinion.

[@ferhaterata4662]

C'mon, this is the best explanation of the topic. Have you read Blitzstein & Hwang 2nd edition, "5.3 Universality of the Uniform"? It is the textbook widely used in US. John covers the topic here extremely intuitive.

[@dlisetteb]

being said: derived distributions implied: derived distribution from an uniform one this is powerfull, but just an introduction

[@mailaddress1185]

are you serious? It was crystal-clear for me same as all the previous lectures