similar to: New car package with new linear.hypothesis function

Hi folks: Does anyone know of a way to do (linear) hypothesis tests of parameters after fitting a maximum-likelihood model w/ optim? I can't seem to find anything like a Wald test whose documentation says it applies to optim output. Also, thanks again to everyone who gave me feedback on the robustness of ML estimation in R! Peter ********************************

I'd like to be able to test linear hypotheses after setting up and running a model using optim or perhaps nlm. One hypothesis I need to test are that the average of several coefficients is less than zero, so I don't believe I can use the likelihood ratio test. I can't seem to find a provision anywhere for testing linear combinations of coefficients after max. likelihood. Cheers

Dear Roman, I've attached the latex file for the copy-edited version of your manuscript -- correcting English grammar and style, and making the paper conform more closely to the R News style. Can you check this over to confirm that I've not altered the meaning of your text and to approve the copy-edits? Thanks, John -------------------------------- John Fox Department of Sociology

On Wed, 4 Jan 2006, Peter Muhlberger wrote: One comment in advance: please use a more meaningful subject. I would have missed this mail if a colleague hadn't pointed me to it. > I'm someone who from time to time comes to R to do applied stats for social > science research. [snip] > I would also prefer not to have to work through a > couple books on R or S+ to learn how to

Dear R-help list members, I forgot to copy my reply to the r-help list. Here's most of it. John -------------------------------- John Fox Department of Sociology McMaster University Hamilton, Ontario Canada L8S 4M4 905-525-9140x23604 http://socserv.mcmaster.ca/jfox -------------------------------- -----Original Message----- From: John Fox [mailto:jfox at mcmaster.ca] Sent: Sunday,

Dear Berwin, Simply stacking the problems and treating the resulting observations as independent will give you the correct coefficients, but incorrect coefficient variances and artificially zero covariances. The approach that I suggested originally -- testing a linear hypothesis using the coefficient estimates and covariances from the multivariate linear model -- seems simple enough. For

Dear all, For analysis of a longitudinal data set with fixed measurement in time I built a gls model (nlme). For testing hypotheses in this model I used the linear.hypothesis function from the car package. A check with the results obtained in SAS proc MIXED with a repeated statement revealed an inconsistency in the results. The problem can be that the linear.hypothesis function (1) only gives the

Dear r-devel list members, I've prepared a version of the Rcmdr package that uses the new tools for internationalization and localization in R 2.1.x. For the present, this is the development version of the Rcmdr package (1.1-0), available at <http://socserv.socsci.mcmaster.ca/jfox/Misc/Rcmdr/index.html>. There is currently a French translation provided with the package, courtesy of

Dear R-devel list members, For several years, I've maintained for my private use and for demonstrations a set of functions [e.g., for indexing, with(), summary(), etc.] that employ the RODBC package and a database to provide a "pseudo" data frame. Row "indexes" can be specified as SQL queries. I know that the idea isn't original, and I haven't done anything with

Dear list members, I've just uploaded version 1.0-0 of the Rcmdr package to CRAN. For people who haven't seen the package before, the "R Commander" provides a basic-statistics graphical user interface to R, based on the tcltk package. The new version incorporates a number of improvements to the R Commander interface, as documented in the CHANGES file distributed with the

Dear list members, I've just uploaded version 1.0-0 of the Rcmdr package to CRAN. For people who haven't seen the package before, the "R Commander" provides a basic-statistics graphical user interface to R, based on the tcltk package. The new version incorporates a number of improvements to the R Commander interface, as documented in the CHANGES file distributed with the

Dear list members, In re-running with GLMM() from the lme4 package a generalized-linear mixed model that I had previously fit with glmmPQL() from MASS, I'm getting a warning of a convergence failure, even when I set the method argument of GLMM() to "PQL": > bang.mod.1 <- glmmPQL(contraception ~ as.factor(children) + cage + urban, + random=~as.factor(children) + cage +

Dear r-help members, Can lmer() in the lme4 package fit models that have a specified within-group correlation structure, as provided, for example, by the correlation argument to lme() in the nlme package? Thanks, John -------------------------------- John Fox Department of Sociology McMaster University Hamilton, Ontario Canada L8S 4M4 905-525-9140x23604 http://socserv.mcmaster.ca/jfox

Dear list members, I've uploaded a new package, called polycor (version 0.5-0), to CRAN. The package has functions for computing polychoric and polyserial correlations, either by maximum-likelihood (in which case, standard errors are available) or by faster approximations. There's also a function to compute "heterogeneous" correlation matrices composed of product-moment,

Dear list members, I've uploaded a new package, called polycor (version 0.5-0), to CRAN. The package has functions for computing polychoric and polyserial correlations, either by maximum-likelihood (in which case, standard errors are available) or by faster approximations. There's also a function to compute "heterogeneous" correlation matrices composed of product-moment,

Dear list members, I think that it's a pity that there's no Windows binary for the rgl package on CRAN. I understand that rgl doesn't compile with the standard Windows tools and that there's a Windows binary on Daniel Adler's web site. But the lack of an rgl binary on CRAN provides presents a potential obstacle to at least some Windows users, such as many students in

Dear r-devel list members, I've experienced the following problem in getting a selection from a Tk list box using tcltk. This is a long-standing problem, but I've finally decided to tackle it. Consider the following: Library(tcltk) top <- tktoplevel() listbox <- tklistbox(top, height="10", width="2", exportselection="FALSE",

Dear list members, I'm encountering the following problem with package dependencies and imports: The Rcmdr package doesn't declare a formal dependency (via depends or suggests in the package DESCRIPTION) on the rgl package because the latter apparently causes problems on some systems. This allows the Rcmdr package to build and check on systems without rgl present; if rgl isn't

Dear R-devel list members, Under the development version of R (2.6.0), I notice that my sem package produces the following warning: checking Rd files ... WARNING Rd files with unknown encoding: residuals.Rd I wonder what the warning means. As far as I can see, residuals.Rd is similar to the other .Rd files in the package. At least for the time being, I won't reproduce residuals.Rd here

Dear R-devel list members, What's the best current advice about writing Fortran code for use in R packages? The Writing R Extensions manual still says that the .Fortran interface is primarily intended for FORTRAN 77 code. In particular, are there portability issues if I use Fortran 95 in a package? For example, I see that Rtools for Windows now include the gfortran compiler. (I know that