Re: teaching statistical methods by rules?
Sorry to get so close to "off topic" but: There are persistent rumours that the U.S. Air Force, which has a massive educational system, including teacher-training, does a far and away better job of "education" than the "public school system." You don't have to be outstandingly intelligent to join the Air Force, either, and it seems to be able to cope with a wide variety of students. a) are the rumours false? b) if not, is the success a function of subject matter rather than their approach to teaching, and hence presumably inapplicable to teaching Statistics? I am not, nor have I ever been, in the Air Force, and hence I am unable to shed any light on the matter, but for DECADES I have been picking up rumours that something about their educational system WORKS. -- "I would predict that there are far greater mistakes waiting to be made by someone with your obvious talent for it." Orac to Vila. [City at the Edge of the World.] ----------- R.W. Hutchinson. | [EMAIL PROTECTED]
Re: algorithm for cumulative t-distribution
>I am trying to write some code which will churn >out a set of values representing the cumulative >density function of the Student t-distribution, >for a given probability and range of DFs. > >Can anyone tell me what I should be trying to >code? > >I can find dozens of tables which list the values >I want, and dozens of applets which generate the >values I want, but I need an algorithm which will >allow me to generate these values on the fly, >given the user's "certainty". > >I need to know which "X value" corresponds to >given "area under the t-distribution". > >Any pointers to papers or URLs, would be much >appreciated. What you are looking for, turns out to be a finite sum of trigonometric functions. They can be found e.g. in the Johnson & Kotz books on "Distributions." -- "I would predict that there are far greater mistakes waiting to be made by someone with your obvious talent for it." Orac to Vila. [City at the Edge of the World.] --- R.W. Hutchinson. | [EMAIL PROTECTED]
Re: Continous multivariate distributions.
>Awhile back, there was a book called Continous Multivariate >Distributions by Samuel Kotz >and others, and published by Wiley. It seems to be out of print. Does >anyone know if there are plans to bring it back into print? Can anyone >recommend any books that will >teach me how to use the multivariate normal distribution? Thanks in >advance. I have been awaiting that volume for some time. It has been delayed for many months at the publisher. The author(s) state that the ball is NOT "in their court." The author(s) also state that "it should not be long now." While a good many volumes have time to come and go out of print in the time that that volume has been "on hold," you have not in fact missed it yet. A little more patience should be rewarded. Meanwhile, if what you need is "some familiarity with the multivariate normal," then there are any number of Multivariate texts which should be able to help you out, without of course the encyclopaedic coverage to be found in the Kotz & Johnson set. If you haven't already tried it, then the multivariate chapter of C.R. Rao's "Linear Statistical Inference" has merit, as do the other chapters. That should offer you "more than an introduction" while of course not covering everything that might be found in some volumes devoted entirely to multivariate applications. The opening chapters on Vectors and Matrices alone, might have been entitled: "everything you ever wanted to know about Linear Algebra but were afraid to ask" and form an excellent foundation for multivariate analysis. Of course, given the DATE on the book - 2nd. ed. 1973 - it can not be denied that SOME useful stuff has been developed since. -- "I would predict that there are far greater mistakes waiting to be made by someone with your obvious talent for it." Orac to Vila. [City at the Edge of the World.] --- R.W. Hutchinson. | [EMAIL PROTECTED]
Re: Looking for probability book.
In <[EMAIL PROTECTED]>, "Patrick D. Rockwell" <[EMAIL PROTECTED]> writes: >There was a book published by Wiley in 1969 called "Discrete >Distributions" by Samuel >Kotz. I tried looking for it at http://www.amazon.com but I couldn't >find the title. >Instead, I found "Univariate Discrete Distributions". Would that be the >same thing but >with a new title? The split the Discrete into two volumes, univariate and multivariate, in the new edition [different publisher, too.] The long awaited final volume to complete the set will be out in another few months. One of the authors conjectures that the publisher, Wiley, wanted a Y2K publication date on it. He can not otherwise explain the delay. [Just as in the first edition, there were two continuous univariate volumes, so the new ed. will be a five volume set when it is complete.] -- "I would predict that there are far greater mistakes waiting to be made by someone with your obvious talent for it." Orac to Vila. [City at the Edge of the World.] --- R.W. Hutchinson. | [EMAIL PROTECTED]
