Mail Archive Search: <+from:bb2@duke.edu>

Results:

References: [ +from:bb2@duke.edu: 23 ]

Total 23 documents matching your query.

1. survReg, frailties, and predictions (score: 1): Author: Brian Beckage <bb2@duke.edu>; Date: Wed, 28 Aug 2002 15:43:04 -0400; I'm trying to predict survival times using a fitted survReg object that includes a gaussian frailty, e.g. fit.final3<-survReg(Herb.surv~Species*Understory+Protection+Overstory*Understory+ frailty(Pl; /archives/html/s-news/2002-08/msg00184.html (7,813 bytes)
2. Calling S from C/C++ (score: 1): Author: Brian Beckage <bb2@duke.edu>; Date: Thu, 29 Aug 2002 14:20:47 -0400; My understanding is that it is NOT possible to call Splus functions (e.g. for random number generation) from C or C++ programs, except in the case where the C/C++ program was originally called from; /archives/html/s-news/2002-08/msg00198.html (6,563 bytes)
3. Re: Calling S from C/C++ (score: 1): Author: Brian Beckage <bb2@duke.edu>; Date: Thu, 29 Aug 2002 16:31:33 -0400; It turns out that you can call Splus from C++ using CONNECT/C++ in Splus 6.0 for Windows and Splus 6.1 for unix. Thanks to all who sent replies. Brian Hi, My understanding is that it is NOT possible; /archives/html/s-news/2002-08/msg00199.html (7,207 bytes)
4. Bessel Function (score: 1): Author: Brian Beckage <bb2@duke.edu>; Date: Thu, 09 May 2002 14:40:06 -0400; Hi everyone, Is there a Splus function for computing a modified Bessel function? I'm using Splus 2000 for Windows. Thanks, Brian; /archives/html/s-news/2002-05/msg00070.html (6,326 bytes)
5. Re: Bessel Function (score: 1): Author: Brian Beckage <bb2@duke.edu>; Date: Thu, 09 May 2002 18:49:57 -0400; Thanks to Leonid Gibiansky for pointing out a function to compute a modified Bessel function. The link is given below. Brian; /archives/html/s-news/2002-05/msg00073.html (7,556 bytes)
6. Contrasts in lme (score: 1): Author: Brian Beckage <bb2@duke.edu>; Date: Sat, 13 Apr 2002 00:26:41 -0400; Hi all, Is is possible to compute contrasts within one factor while holding the level of a second factor constant (using lme)? For instance, considering the oats dataset...how could we construct a co; /archives/html/s-news/2002-04/msg00105.html (7,229 bytes)
7. Matrix question (score: 1): Author: Brian Beckage <bb2@duke.edu>; Date: Fri, 04 Jan 2002 11:06:18 -0500; If, for example, I have a 3 by 3 matrix called 'tmp' that contains 1:9, is there a simple way (that avoids a 'for' loop) of retrieving the row and column number of a given entry? Something of the fo; /archives/html/s-news/2002-01/msg00028.html (6,518 bytes)
8. Summary: Matrix question (score: 1): Author: Brian Beckage <bb2@duke.edu>; Date: Fri, 04 Jan 2002 13:50:32 -0500; Thanks to everyone who responded to my original question (copied below). I've summarized the responses below in no particular order. Brian is there a simple way (that avoids a 'for' loop) of retrievi; /archives/html/s-news/2002-01/msg00032.html (8,707 bytes)
9. arima.mle() and AIC (score: 1): Author: Brian Beckage <bb2@duke.edu>; Date: Mon, 14 Jan 2002 17:14:48 -0500; Should centering a covariate used in an ARMA model change the value of AIC or the AR or MA coefficients? Below, I fit an AR(1) model to the same data but using a centered (mean subtracted from each; /archives/html/s-news/2002-01/msg00089.html (8,809 bytes)
10. Double Integration (score: 1): Author: Brian Beckage <bb2@duke.edu>; Date: Sat, 17 Nov 2001 12:40:47 -0500; I need to integrate the following function in two dimensions: fct<-function(x, y){ (a^-b)*b*exp(s*x-(((x^2 + y^2)^0.5)/a)^b)*(x^2 + y^2)^(0.5*(b-1))} over x (0, Inf) and y (-Inf, Inf). a,b, and s ar; /archives/html/s-news/2001-11/msg00156.html (9,390 bytes)
11. Design matrix not full rank (score: 1): Author: Brian Beckage <bb2@duke.edu>; Date: Sun, 14 Oct 2001 21:18:14 -0500; Hi everyone, I have a large design matrix (2465 rows by 26 columns) that is not full rank, i.e. when I use bic.logit(), which calls glim(), I get an error message to this effect. The design matrix co; /archives/html/s-news/2001-10/msg00095.html (7,114 bytes)
12. stepAIC and survReg (score: 1): Author: Brian Beckage <bb2@duke.edu>; Date: Tue, 06 Mar 2001 09:50:39 -0600; Hi all, Is it possible to use stepAIC with survReg? I currently receive the following message--"Error in UseMethod("extractAIC"): No specific or default method found". Any suggestions as to how to so; /archives/html/s-news/2001-03/msg00044.html (6,463 bytes)
13. stepAIC and survReg again (score: 1): Author: Brian Beckage <bb2@duke.edu>; Date: Tue, 06 Mar 2001 09:58:55 -0600; I forgot to mention that I'm using Splus 2000 R3. Brian; /archives/html/s-news/2001-03/msg00046.html (6,139 bytes)
14. Overlaying Plots (score: 1): Author: Brian Beckage <bb2@duke.edu>; Date: Wed, 06 Dec 2000 10:12:49 -0600; I would like to plot two time series on the same plot but with different scales for each of the series, i.e. one scale on the left hand y-axis and the other on the right hand y-axis. Can this be don; /archives/html/s-news/2000-12/msg00045.html (6,291 bytes)
15. Summary: Overlaying Plots (score: 1): Author: Brian Beckage <bb2@duke.edu>; Date: Wed, 06 Dec 2000 12:46:57 -0600; Thanks to all who responded! My original posting is at the bottom of this message. Matthew Austin and Stephen Smith both suggested using subplot (see 311--314 of the Programmer's Guide). Sean Keenan; /archives/html/s-news/2000-12/msg00056.html (7,739 bytes)
16. [S] Simple graphics question (score: 1): Author: Brian Beckage <bb2@duke.edu>; Date: Mon, 06 Mar 2000 10:36:38 -0500; I migrated from Splus 5.1 to Splus 2000, r2. I'm accustomed to creating all of my graphics using the command window. However, I would like to directly edit my plots using all of the 'point and click; /archives/html/s-news/2000-03/msg00057.html (7,950 bytes)
17. [S] Summary: Simple graphics question (score: 1): Author: Brian Beckage <bb2@duke.edu>; Date: Mon, 06 Mar 2000 12:24:05 -0500; Thanks to replies from Andy Liaw, George Bagnall, Daniel Gadd, and Prof Brian Ripley. It is as simple as checking the "Create Editable Graphics" option which it seems can be done in a number of ways.; /archives/html/s-news/2000-03/msg00059.html (9,431 bytes)
18. [S] Problem in kaplanMeier (score: 1): Author: Brian Beckage <bb2@duke.edu>; Date: Mon, 24 Jan 2000 01:26:40 -0500; kaplanMeier seems to be misidentifying which survivor functions belong to which levels of a factor (when using the formula argument; and I'm using Splus 2000 release 1). For instance, data=Surv.fact; /archives/html/s-news/2000-01/msg00168.html (10,574 bytes)
19. [S] survreg question (score: 1): Author: Brian Beckage <bb2@duke.edu>; Date: Fri, 31 Dec 1999 00:54:38 -0500; I'm fitting a Weibull model to interval (and right) censored survival data using survreg in Splus 5.1. I'm trying to use the "strata" argument in the model formula as described in Venables and Riple; /archives/html/s-news/1999-12/msg00191.html (9,642 bytes)
20. [S] Subsetting Vectors (score: 1): Author: Brian Beckage <bb2@duke.edu>; Date: Fri, 16 Apr 1999 23:24:49 -0400; I have a simple question regarding the subsetting of vectors (or matrices). Given a vector test<-c(1,2,1,1,3,4,5,8,9,8) I can replace the value of a subset of it using: test[test>3]<-0 resulting in [; /archives/html/s-news/1999-04/msg00156.html (7,394 bytes)

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