Re: WHY IS SOUND to get C.V.´s from Mont Carlo
- From: "Luis A. Afonso" <licas_@xxxxxxxxxxx>
- Date: Fri, 30 Mar 2007 00:02:49 EDT
Jack Tomsky wrote
*** Three times in your post, you stated that you were generating data from N(0,1) or from N(m,1). What are the odds that you would make the same typo three times in a row? I am such a decent person that I take what you write at face value.
Why waste time doing Monte Carlo simulations when there exist exact analytical solutions? For example, confidence limits on quantiles can be obtained from the noncentral t distribution. Confidence limits on the normal mean can be obtained from the normal (if the standard deviation is known) or the t (if the standard deviation is unknown). Confidence limits on the standard deviation from a normal can be obtained from the chi-square. Confidence limits on the exponential parameter can be obtained from the chi-square.
Jack ***
My response
BEFORE you decide whether YES or NO Monte Carlo simulations are able to get valuable critical values (and confidence intervals) which is the MATTER THAT MATTERS (and the threads name) I do not answer your questions. They shall not be forgotten, be sure, I will respond as long as you make you duty, be sure.
________licas (Luis A. Afonso)
.
- Prev by Date: Re: WHY IS SOUND to get C.V.´s from Mont Carlo
- Next by Date: Re: A letter to a statistician. Emailed and airmailed and no reply
- Previous by thread: Re: WHY IS SOUND to get C.V.´s from Mont Carlo
- Next by thread: Re: WHY IS SOUND to get C.V.´s from Mont Carlo
- Index(es):
Relevant Pages
|
|
