Revision as of 15:25, 14 May 2014 by Rhea (Talk | contribs)

Comments for: The principles for how to generate random samples from a Gaussian distribution

A slecture by Joonsoo Kim


Please leave me comment below if you have any questions, if you notice any errors or if you would like to discuss a topic further.


Review by Khalid Tahboub:

Great job! few minor remarks:

1) I think the first equation should be

$ F^{-1}(u)=inf\{ x|F(x)\geq u, \quad u\in [0, 1] \} $

instead of

$ F^{-1}(u)=inf\{ x|F(x)\leq u, \quad u\in [0, 1] \} $
2) How we reach
$ X <- F^{-1}(U)\quad $
from
$ F^{-1}(u)=inf\{ x|F(x)\geq u, \quad u\in [0, 1] \} $
is not very clear to me


3)I think the equation

$ F(x) = \int_{-\infty}^x \lambda exp(-\lambda x') dx' = \int_0^x \lambda exp(-\lambda x') dx' = [-exp(-\lambda x')]_0^x = 1-exp(-\lambda x) \leq u $

should be instead

$ F(x) = \int_{-\infty}^x \lambda exp(-\lambda x') dx' = \int_0^x \lambda exp(-\lambda x') dx' = [-exp(-\lambda x')]_0^x = 1-exp(-\lambda x) \geq u $

4) I think it might give a nice demonstration if you plot the histogram of U and X to show that this method really functions in the desired way.


Write Question/Comment Here


Write Question/Comment Here




Back to The_principles_for_how_to_generate_random_samples_from_a_Gaussian_distribution

Alumni Liaison

Abstract algebra continues the conceptual developments of linear algebra, on an even grander scale.

Dr. Paul Garrett