I have a function that has worked in the past but for some reason no longer works. The function searches for combinations of interactions that might be useful in a logistic regression model. It now g
I'd like to name numerous objects based on an argument in a function. An example of how I'd do it in SAS is given below . %MACRO UPDATE(OLDMONTH,NEWMONTH); program1&NEWMONTH = some modification of pr
Assign( ) and get( ) work perfectly! Thanks James. --Original Message-- From: james.holtman@convergys.com [mailto:james.holtman@convergys.com] Sent: Friday, January 07, 2005 10:57 AM To: Russell.Ivor
I’d like to name numerous objects based on an argument in a function. An example of how I’d do it in SAS is given below .<o:p></o:p> <![if !supportEmptyParas]> <![endif]><o:p></o:p> %MACR
I have a function that sometimes encounters a singular matrix in a for loop. I believe I can keep the loop going with the try( ) function (I'm still working on it), but I also want to keep a true fal
Many thanks to Sundar Dorai-Raj, Bill Dunlap, Tony Plate, and John James for ideas and explanations. It seems the error.check( ) function uses features that no longer work. The simplest solution that
Hi S-Plus group. I have the following code that works; but there has to be a more efficient way of doing it, especially if there are many levels of either or both factors. Basically the code says tha
Many thanks to Matt Austin, Bill Dunlap, Thomas Jagger, Kristina Kollen, Julian Taylor, and Henrik Nielsen for the numerous solutions to my problem. I found the simplest after all was grade2004 <- if
Hi group. Does anyone know how to get expected counts in each cell of the crosstabs function ? I can generate the expected counts as a separate matrix and would like to 'add an attribute' labeled "Ex
I want to run the Function( ) function along with the sascode( ) function but I'm getting the following error: Problem in parse(text = z): End of file before closing "}" on input line 1 The model obj
Hi Group. I'd like to convert each vector of a data frame to unit vectors. I couldn't find a function to do it so I'm trying to write one and have a small data frame (10x3) to work with. I'm having t
I'm trying to write a function that would use Greek letters in a title and legend on a plot. It seems to only way I can use Greek letters is by using the guiPlot( ) rather than just plot( ). My probl
Many thanks to David L Lorenz, Bert Gunter, Stephen Smith, Alan Zaslavsky, and Spencer Graves for very quick and very helpful responses. I found the best fix in my case was to use the assign( ) funct
I have a model testmodel <- lrm(depvar ~ age+fscore+accts+lates+stage) As part of my diagnostics I want to manipulate things one continuous measure at a time. I want to write a function to run a new
I have two questions with using the T/F logic of arima.mle$"converged" (1) It seems that S-Plus is reversing the logic in the following statements: [1] F [1] F [1] 5 If the model did not converge and
Thanks to Bill Dunlap, Pierre Duchesne, and Sundar Dorai-Raj the solution is of course a double ==. I knew that; duh! It's been one of those weeks. Thanks for keeping me straight I have two questions
Author: Russell Ivory <Russell.Ivory@MerrickBank.com>
Date: Wed, 5 Mar 2003 08:30:01 -0700
I have a function that generates a table of simple logistic regression coefficients for a list of potential independent variables. The function has the form function(depvar,...,alpha=.05) where ... i
Author: Russell Ivory <Russell.Ivory@MerrickBank.com>
Date: Wed, 5 Mar 2003 15:44:23 -0700
Thanks to all for the suggestions. The solution (from robbing parts of data.frame) was dots <- match.call(expand.dots = F)$... to capture the list of dependent variables, and c(unlist(dots)[-1]) as p