search for: firths

Dear R-Help List, I'm trying to implement Firth's (1993) bias correction for log-linear models. Firth (1993) states that such a correction can be implemented by supplementing the data with a function of h_i, the diagonals from the hat matrix, but doesn't provide further details. I can see that for a saturated log-linear model, h_i=1 for all i, hence one just adds 1/2 to each count,

> ---------------------------------------------------------------------- > > Message: 1 > Date: Wed, 9 Jul 2003 10:53:27 +0100 > From: David Firth <david.firth at nuffield.oxford.ac.uk> > Subject: [R] packaged datasets in .csv format > To: r-help at stat.math.ethz.ch > Message-ID: > <307D34CE-B1F3-11D7-A8D2-0050E4C03977 at nuffield.oxford.ac.uk> >

Is there a straightforward way to convert character information from decimal representation to the octal one used by R? I'd like something like a function ascii(number,base=10), such that > ascii(91) [1] "\133" I can easily do the mapping from 91 to 133, but what is a good way to operate on 133 to deliver "\133"? Would a lookup table be a better solution? David

hi all, I have a binary data set and am now confronted with a "separation" issue. I have two predictors, mood (neutral and sad) and game type (fair and non-fair). By "separation", I mean that in the non-fair game, whereas 20% (4/20) of sad-mood participants presented a positive response (coded as 1) in the non-fair game, none of neutral-mood participants did so (0/20). Thus,

I am making my first attempts at using some Fortran code with R, and so far it's going OK. To print from my Fortran programs, it seems I need subroutines dblepr, realpr and intpr. From the excellent "Writing R Extensions" document: "Three subroutines are provided to ease the output of information from FORTRAN code. subroutine dblepr(label, nchar, data, ndata)

I am having a little problem in relation to sending output to the R Console window, when a function is called by pushing a button in a Tk window. Actually there seem to be two components to the problem: -- the R prompt does not re-appear until <return> is pressed -- text output to the R Console window *disappears* after <return> is pressed The following, in R 1.1.1 under

No, blas/veclib is tested, so aprt this extreme case you should report some other more commonly used cases in which something fails on OS X. This will help us to work it out. As said, I'll try some tests without using veclib and let you know. I've fowarded this mail to r-devel, which seems to be the right place, so for future msg on the subject please use r-devel. stefano On

Hi List I am hoping that someone here could perhaps give me a straight answer on a question that someone asked me today I have always belived that if you have 5 hard drives 1 50gb second 50gb third 20gb fourth 60gb firth 30gb that the largest would then be the size of the smallest disk, not 80 or 100 or 120 for that matter or am I wrong here? Regards? Per Qvindesland? -------------- next part

Is there some neat way of storing the printed representation of an object as a character string? I can see how it could be done via disk using sink() and then scan(), but that's ugly. Something like Lisp-Stat's "with-output-to-string" macro perhaps? David Firth Phone +44 1865 278544 Nuffield College Fax +44 1865 278621 Oxford OX1

Hi David, I hope all is well w/ you. Co-operation beetwen R and Ox would be most welcome. A possible problem lies in the fact that, although free for academic use, Ox is not open source. I too have been using Ox for computer-intensive programming. I also code some of my programs in C, but I always try Ox first, and in most cases it is efficient enough. For details, see

I propose the following enhancements and changes to plot.lm(), the most important of which is the addition of a Residuals vs Leverage plot. (1) A residual versus leverage plot has been added, available by specifying which = 5, and not included as one of the default plots. Contours of Cook's distance are included, by default at values of 0.5 and 1.0. The labeled points, if any, are those

I suppose this could be described as a feature (it seems to be similar in S-Plus), but it looks to me more like a bug. Why can't the assignment below to a row of "emptyframe" (or "anotherframe") be made? This with R --vanilla (version info below). Regards -- David David Firth Phone +44 1865 278544 Nuffield College Fax +44 1865

Full_Name: David Firth Version: 2.0.0 OS: Mac OS 10.3.5 Submission from: (NULL) (81.178.234.156) When running R in R Cocoa GUI 1.0 (v2004-10-14), I have > Sys.getenv()[["R_LIBS"]] [1] "/Users/david/Library/R" > .libPaths() [1] "/Users/david/Library/R/library" [2] "/Library/Frameworks/R.framework/Resources/library" The pasting of

I'm writing a compiler using LLVM, and suddenly today I started to get this error, when emitting to a .s file. Assertion failed: (I != M+Size && I->FromReg == RegNum && "Invalid RegNum"), function getLLVMRegNum, file MCRegisterInfo.cpp, line 78. I'm emitting using x86_64-apple-darwin. Any thoughts about what could be causing this? LLVM validation does not

I am aware of the nice R package "logspline", which does smooth density estimation from interval-censored data (that is, values that are known to lie in a specified interval rather than known exactly). Function logspline.fit uses a maximum penalized likelihood method, with the penalty related to the number of knots used in a cubic regression-spline fit. I need to be able to do some

Full_Name: David Firth Version: 1.3.1 OS: linux and others Submission from: (NULL) (163.1.103.121) help(lm) says that An object of class `"lm"' is a list containing at least the following components: ... formula: the formula supplied. But is the "formula" component necessarily present? An example: > x<-rnorm(10) > y<-rnorm(10) >

Full_Name: David Firth Version: 2.1.1 OS: Mac OS X Submission from: (NULL) (137.205.240.25) A small inconsistency in the "Writing R Extensions" manual: at http://cran.r-project.org/doc/manuals/R-exts.html#The-DESCRIPTION-file the displayed example has Title: My first collection of functions but the text that follows it says "The Title field...should be capitalized",

Full_Name: David Firth Version: 1.8.0 OS: Mac OS 10.2.8 Submission from: (NULL) (80.177.12.48) In .Rd files, we use \url{stuff} to refer to a URL on the web. Just as the documentation says. Perhaps this isn't a bug, then, but certainly it's a trap for the unwary. If the construction \link{url{something}} appears in a .Rd file, then Rdconv --type=html ... gets stuck and appears

I'm writing a JIT for my toy language, and I'm confused by one thing. How do I, with IR Builder, Insert a pointer constant? For the first version of my jit, I really only need LLVM to pass around opaque types and then pass those variables to C functions that return other opaque types. So for instance, I have a Int object that wraps the GNU GMP routines. A want to be able to tell IRBuilder

Hi, I got two questions and would really appreciate any help from here. First, is the penalized maximum likelihood estimation(Firth Type Estimation) only fit for binary response (0,1 or TRUE, FALSE)? Can it be applied to multinomial logistic regression? If yes, what's the formula for LL and U(beta_i)? Can someone point me to the right reference? Second, when I used *logistf *on a dataset with