Mail Archive Search: <+from:jones@reed.edu>

Results:

References: [ +from:jones@reed.edu: 29 ]

Total 29 documents matching your query.

1. independent variables (score: 1): Author: Albyn Jones <jones@reed.edu>; Date: Tue, 30 Oct 2001 10:29:30 -0800; unfortunately I deleted the original message about polynomials in glm's, so this is possibly missing the point of the question... it appeared to me that the question derived from the confusion caused; /archives/html/s-news/2001-10/msg00209.html (7,100 bytes)
2. Re: Reference Books for Midxed Distribution (score: 1): Author: Albyn Jones <jones@reed.edu>; Date: Wed, 7 Mar 2001 14:40:16 -0800 (PST); Titterington, D. M.;Smith, A. F. M.;Makov, U. E. Statistical analysis of finite mixture distributions 85Wiley Everitt, B. S.;Hand, D. J. Finite mixture distributions 81ChapmnHall albyn -- http://www.; /archives/html/s-news/2001-03/msg00059.html (7,484 bytes)
3. Re: [S] Built-in function for higher moments? (score: 1): Author: .com>; Date: Tue, 22 Aug 2000 08:49:08 -0700 (PDT); as to how the .First and .First.local() functions differ. The S-Plus documentation is vague on specifics (only saying that .First.local() could be used to; /archives/html/s-news/2000-08/msg00185.html (8,894 bytes)
4. Re: [S] Built-in function for higher moments? (score: 1): Author: c.uk>; Date: Wed, 23 Aug 2000 10:15:39 -0700 (PDT); near model. Yes, no. You do need to be very careful in interpreting unbalanced anovas, though. -- Brian D. Ripley, ripley@stats.ox.ac.uk Professor of Appli; /archives/html/s-news/2000-08/msg00204.html (8,904 bytes)
5. Re: [S] How can I fit a t-distribution to a sample? (score: 1): Author: Albyn Jones <jones@reed.edu>; Date: Wed, 31 May 2000 10:08:09 -0700 (PDT); li; /archives/html/s-news/2000-05/msg00300.html (8,732 bytes)
6. Re: [S] "Copula" (score: 1): Author: Albyn Jones <jones@reed.edu>; Date: Tue, 11 Jan 2000 11:54:12 -0800 (PST); see Genest, Christian;MacKay, Jock The joy of copulas: Bivariate distributions with uniform marginals 86AmerStat 40 280- 283 J Lee, A. J. Generating random binary deviates having fixed marginal distr; /archives/html/s-news/2000-01/msg00054.html (8,952 bytes)
7. Re: [S] Splus in mixed architecture environment? (score: 1): Author: Albyn Jones <jones@reed.edu>; Date: Thu, 9 Dec 1999 12:07:39 -0800 (PST); I have been living with this problem for several years. My simple solution allowing a single home directory is to have a subdirectory for S for the non-default machine. I occasionally forget where I; /archives/html/s-news/1999-12/msg00066.html (8,843 bytes)
8. Re: [S] polygons intersection (score: 1): Author: Albyn Jones <jones@reed.edu>; Date: Mon, 15 Nov 1999 08:34:13 -0800 (PST); How are your polygons given? Do you have the vertices, the inequalities for the intersecting half planes, or some specification of the faces as equations? albyn -- http://www.reed.edu/~jones Albyn Jo; /archives/html/s-news/1999-11/msg00148.html (7,968 bytes)
9. d deviation estimates (score: 1): Author: Albyn Jones <jones@reed.edu>; Date: Mon, 29 Nov 1999 15:59:34 -0800 (PST); I apologize for asking a non-S question here, but still hope someone has the answer.... I have a dataset that I have often used as an example for logistic regression: Age N Blind 1 20 50 6 2 35 50 17; /archives/html/s-news/1999-11/msg00286.html (7,595 bytes)
10. aints using predict() (score: 1): Author: Albyn Jones <jones@reed.edu>; Date: Tue, 30 Nov 1999 10:48:12 -0800 (PST); Thanks to Bill Dunlap, Jeff Simonoff, Brian Ripley, and at least one other good soul who responded to my query about the "Kalythos" dataset. All seem to agree that it is probably made up data, though; /archives/html/s-news/1999-11/msg00296.html (7,594 bytes)
11. Re: [S] Sex vs. gender. (score: 1): Author: Albyn Jones <jones@reed.edu>; Date: Sat, 9 Oct 1999 21:00:49 -0700 (PDT); I doubt that is the issue at all. I suspect it has more to do with the tendency in the social sciences to speak in terms of things like "gender roles" as opposed to "sex roles", with the notion that; /archives/html/s-news/1999-10/msg00080.html (9,215 bytes)
12. Re: [S] Sex/Gender (score: 1): Author: Albyn Jones <jones@reed.edu>; Date: Mon, 11 Oct 1999 19:46:50 -0700 (PDT); Ok, I've seen this often enough now I just have to respond, in spite of the fact that its really straying from the S-news charter, and I don't know beans about linguistics... Gender is a technical te; /archives/html/s-news/1999-10/msg00095.html (8,819 bytes)
13. Re: [S] bug in histogram ? (score: 1): Author: Albyn Jones <jones@reed.edu>; Date: Sat, 30 Oct 1999 13:55:15 -0700 (PDT); a histogram, or in general a density does not have to be bounded by 1, the total area has to be 1. Consider for example the normal with small standard deviation: [1] 398.9423 albyn -- http://www.reed; /archives/html/s-news/1999-10/msg00241.html (8,642 bytes)
14. Re: [S] Confidence interval for a discrete Poisson variable with normally distributed mean (score: 1): Author: Albyn Jones <jones@reed.edu>; Date: Tue, 28 Sep 1999 08:38:04 -0700 (PDT); A more natural mixture model for the Poisson is the Gamma mixture (Poisson mean follows a Gamma distribution). This will automatically satisfy the positivity constraint... A Gamma mixture of Poissons; /archives/html/s-news/1999-09/msg00267.html (8,782 bytes)
15. Re: [S] Introductory level Stat courses based on S-PLUS. (score: 1): Author: Albyn Jones <jones@reed.edu>; Date: Sat, 19 Jun 1999 11:02:58 -0700 (PDT); I have been using S+ (and before that old S) for introductory courses since 1986. I have students from all areas in the course - no stat for bio, stat for econ, etc. How successfully? Good question :; /archives/html/s-news/1999-06/msg00157.html (9,531 bytes)
16. RE: [S] Manuals (score: 1): Author: Albyn Jones <jones@reed.edu>; Date: Wed, 12 May 1999 15:15:14 -0700 (PDT); perhaps part of the infamous silent majority - this is a natural situation for self selection bias. I haven't looked at a hardcopy S manual (OK, I'm not counting things like V&R as a manual) for year; /archives/html/s-news/1999-05/msg00128.html (9,390 bytes)
17. Re: [S] sum of matrix col. (score: 1): Author: Albyn Jones <jones@reed.edu>; Date: Tue, 6 Apr 1999 14:58:39 -0700 (PDT); I thought this shoudl be an easy application of cumsum and apply, but it turned out to be slightly tricky for reaasons I don't understand yet. in Version 3.4 Release 1 for Sun SPARC, SunOS 5.3: [,1]; /archives/html/s-news/1999-04/msg00031.html (10,930 bytes)
18. Re: [S] Diagonal matrix multiplication (score: 1): Author: Albyn Jones <jones@reed.edu>; Date: Fri, 16 Apr 1999 14:02:11 -0700 (PDT); [,1] [,2] [,3] [1,] 1 5 9 [2,] 2 6 10 [3,] 3 7 11 [4,] 4 8 12 [,1] [,2] [,3] [1,] 0.5 5 18 [2,] 1.0 6 20 [3,] 1.5 7 22 [4,] 2.0 8 24 [,1] [,2] [,3] [1,] 0.5 5 18 [2,] 1.0 6 20 [3,] 1.5 7 22 [4,] 2.0; /archives/html/s-news/1999-04/msg00154.html (9,156 bytes)
19. Re: [S] bump hunting (score: 1): Author: jones@reed.edu (Albyn Jones); Date: Wed, 10 Feb 1999 10:19:52 -0800 (PST); one natural alternative is to fit a mixture distribution. as I understand the problem you have several peaks in a histogram, and want to fit something other than a single parametric distribution. if; /archives/html/s-news/1999-02/msg00050.html (9,279 bytes)
20. msg00059.html (score: 1): Author: jones@reed.edu (Albyn Jones); Date: Thu, 11 Feb 1999 10:04:19 -0800 (PST); on a sun sparc it works fine: Version 3.4 Release 1 for Sun SPARC, SunOS 5.3 : 1996 [1] 0 0 0 0 0 1 0 0 0 2 0 0 0 0 0 0 0 0 0 0 One-sample Kolmogorov-Smirnov Test; hypothesized distribution = geometr; /archives/html/s-news/1999-02/msg00059.html (8,287 bytes)

Current List: 1 - 20
Page: [1] [2]