Search String: Display: Description: Sort:

Results:

References: [ +from:William.Venables@cmis.CSIRO.AU: 34 ]

Total 34 documents matching your query.

1. *****SPAM***** Re: important excel document (score: 1)
Author: william.venables@cmis.csiro.au
Date: Mon, 5 Apr 2004 11:09:12 +0800
Please read the important document. ** NOTE: An attachment was deleted from this part of the message, because it failed one or more checks by the virus scanning system. The file has been quarantined
/archives/html/s-news/2004-04/msg00010.html (6,825 bytes)

2. Re: t-test for binary data (score: 1)
Author: Bill Venables <William.Venables@cmis.CSIRO.AU>
Date: Fri, 03 Nov 2000 14:11:16 +1000
Yes, finnicky. If you have large samples it doesn't matter at all and if you have small samples you are in a spot of bother no matter what you do... It really is not worth starting that non-producti
/archives/html/s-news/2000-11/msg00018.html (8,685 bytes)

3. Re: contrasts for fully crossed anova (score: 1)
Author: Bill Venables <William.Venables@cmis.CSIRO.AU>
Date: Wed, 11 Oct 2000 16:26:41 +1000
Martin H H Stevens asks: This is the contrast matrix that will be used unless you supply something else. It's very hard to know what you mean, but I guess it goes something like this: "If I fit a two
/archives/html/s-news/2000-10/msg00101.html (10,051 bytes)

4. Re: troubles with ms() (score: 1)
Author: Bill Venables <William.Venables@cmis.CSIRO.AU>
Date: Fri, 13 Oct 2000 09:57:03 +1000
Aaron J Mackey reports: well done...so did I! ** Look carefully at the double-quote signs in the line above !!!! ** Indeed. (I'm never quite sure how to handle these completely global questions...) T
/archives/html/s-news/2000-10/msg00131.html (10,424 bytes)

5. [S] CSIRO Jobs in Marine Ecosystem Modelling in Australia (score: 1)
Author: Bill Venables <William.Venables@cmis.CSIRO.AU>
Date: Sun, 3 Sep 2000 12:37:31 +1000 (EST)
There are two positons available here in the Tropical and Pelagic Ecosystems Program at the CSIRO Marine Research Laboratories, Cleveland, Queensland, Australia. The official ads appeared in the 2nd
/archives/html/s-news/2000-09/msg00011.html (9,055 bytes)

6. Re: [S] Inverting non-square matrices (score: 1)
Author: Bill Venables <William.Venables@cmis.CSIRO.AU>
Date: Fri, 08 Sep 2000 16:12:30 +1000
...make that t(x) %*% solve(crossprod(t(x))) I think. -- Bill Venables, Statistician, CMIS Environmetrics Project CSIRO Marine Labs, PO Box 120, Cleveland, Qld, AUSTRALIA. 4163 Tel: +61 7 3826 7251 E
/archives/html/s-news/2000-09/msg00058.html (8,232 bytes)

7. Re: [S] answers to trying to avoid looping (score: 1)
Author: Bill Venables <William.Venables@cmis.CSIRO.AU>
Date: Sat, 16 Sep 2000 13:31:18 +1000
I regret to say I have to send my last message again as it accidently became polluted by MicroSoft junk. The most pernicious garbling was to replace the innocent <- by the egregious <<- globally. Now
/archives/html/s-news/2000-09/msg00116.html (10,811 bytes)

8. Re: [S] statistics question (score: 1)
Author: .com>
Date: Thu, 24 Aug 2000 12:44:38 +1000
ature, is "it depends". Mainly, it depends on your how much virtual memory (swap space) and physical RAM you have. As long as the operating system will pro
/archives/html/s-news/2000-08/msg00210.html (9,604 bytes)

9. Re: [S] Model syntax for split plot designs (score: 1)
Author: Bill Venables <William.Venables@cmis.CSIRO.AU>
Date: Sat, 3 Jun 2000 17:52:33 +1000 (EST)
Rolf Turner gets very excited over S-PLUS anova in the following terms: Steady there Rolf, you'll do yourself a damage... :-) It's a cross-Atlantic culture clash. The S-PLUS style is much more like G
/archives/html/s-news/2000-06/msg00023.html (21,347 bytes)

10. Re: [S] deriv versus deriv3 (score: 1)
Author: Bill Venables <William.Venables@cmis.CSIRO.AU>
Date: Thu, 24 Feb 2000 15:35:14 +1000
deriv3() may be in the MASS library now, but it is there by kind permission of the author, David M. Smith, now of MathSoft UK. It is a bug that was introduced into S-PLUS 4.x and has remained safely
/archives/html/s-news/2000-02/msg00234.html (10,523 bytes)

11. Re: [S] deriv versus deriv3 (score: 1)
Author: Bill Venables <William.Venables@cmis.CSIRO.AU>
Date: Fri, 25 Feb 2000 11:15:00 +1000
Doug Bates cries "foul!" in the following terms: No, I don't think so. I must admit to a tremendous sense of "deja-vu all over again" when I saw this question, but the full details did not come back
/archives/html/s-news/2000-02/msg00253.html (13,640 bytes)

12. Re: [S] Fast looping (score: 1)
Author: Bill Venables <William.Venables@cmis.CSIRO.AU>
Date: Tue, 11 Jan 2000 09:02:44 +1000
I think so. n <- 10^4 px <- 1/10 x <- rbinom(n, 1, px) py <- 1 - eta + x*(theta - 1 + eta) xs <- rbinom(n, 1, py) -- -- Bill Venables, Statistician, CMIS Environmetrics Project. Physical address: Po
/archives/html/s-news/2000-01/msg00047.html (8,677 bytes)

13. Re: [S] "Copula" (score: 1)
Author: Bill Venables <William.Venables@cmis.CSIRO.AU>
Date: Wed, 12 Jan 2000 09:39:58 +1000
This idea is gaining more attention than I would have thought it was really worth so perhaps we (ie mainly I) should be a bit more aware of it. If you take any univariate random variable X with cont
/archives/html/s-news/2000-01/msg00056.html (10,752 bytes)

14. Re: [S] deriv3() question (score: 1)
Author: Bill Venables <William.Venables@cmis.CSIRO.AU>
Date: Sat, 15 Jan 2000 09:13:14 +1000
Ming Ji asks: The answer is "no". deriv() and deriv3() are symbolic differentiators that work with a rather limited range of expressions usually specified by an S formula. For it to work with a new f
/archives/html/s-news/2000-01/msg00089.html (9,758 bytes)

15. Re: [S] Finding most closely matching element (score: 1)
Author: Bill Venables <William.Venables@cmis.CSIRO.AU>
Date: Tue, 18 Jan 2000 00:53:29 +1000 (EST)
Frank E Harrell Jr <fharrell@virginia.edu> posed the question: Curiously only recently I had a two-dimensional version of precisely this problem, which I now propose as a challenge for anyone with a
/archives/html/s-news/2000-01/msg00104.html (10,152 bytes)

16. Re: [S] data.dump from Splus5 to Splus 2000 on Windows (score: 1)
Author: Bill Venables <William.Venables@cmis.CSIRO.AU>
Date: Sat, 22 Jan 2000 16:30:52 +1000
I have also had problems of this kind going from S-PLUS 5.1 release 2 on Sun SPARC solaris 7 to S-PLUS 2000. I believe the problem lies with data.dump on S-PLUS 5.1 which seems to be seriously broke
/archives/html/s-news/2000-01/msg00163.html (9,544 bytes)

17. Re: [S] on vcov() (score: 1)
Author: Bill Venables <William.Venables@cmis.CSIRO.AU>
Date: Tue, 25 Jan 2000 00:36:13 +1000
Dear Stephanie, As Professor Ripley has already explained, the variance matrix returned by vcov for the coefficient vector estimated in the linear model. For factor models these are a reduced set co
/archives/html/s-news/2000-01/msg00174.html (10,724 bytes)

18. Re: [S] confint / Error in profile.nls (score: 1)
Author: Bill Venables <William.Venables@cmis.CSIRO.AU>
Date: Wed, 26 Jan 2000 09:29:56 +1000
Kevin Wright asks: profile.nls is not part of the MASS library, of course, it is part of S-PLUS. profile.glm is part of the MASS library and it has a trace= argument that allows the profiling to be t
/archives/html/s-news/2000-01/msg00188.html (10,655 bytes)

19. Re: [S] thanks for bivariate...replies (score: 1)
Author: Bill Venables <William.Venables@cmis.CSIRO.AU>
Date: Wed, 22 Dec 1999 09:43:33 +1000
to which Brian Ripley replied: and I confirm: Absolutely! The original, still the best and open source to boot! Bill. -- -- Bill Venables, Statistician, CMIS Environmetrics Project. Physical address:
/archives/html/s-news/1999-12/msg00159.html (8,485 bytes)

20. Re: [S] Testing between GAMs (score: 1)
Author: Bill Venables <William.Venables@cmis.CSIRO.AU>
Date: Tue, 02 Nov 1999 13:05:17 +1000
There are a couple of points to make here. The "Negative Binomial ala MASS" is not a GLM, comparisons between negative binomial models are done using likelihood ratio tests and the true log-likeliho
/archives/html/s-news/1999-11/msg00007.html (11,321 bytes)

Current List: 1 - 20
Page: [1] [2]
This search system is powered by Namazu