I recall seeing the same phenomenon (predict.glm producing answers on a different scale for negbin objects than glm objects with the same type parameter) a couple of years ago with earlier versions o
I've been inserting arbitrary symbols into my windows-created graphs for some time now. If you're using windows, you can insert greek, etc., in the axis labels using the GUI (Insert|Symbol...). The a
I'd like to thank everyone for their responses regarding my multiple-NA problem. The responses universally pointed to the well-known problem in S-PLUS regarding confusing NaN's with NA's in certain s
S-PLUS 2000 R3 on a Dell Wintel box running W2000. I've run across something very obscure that I could use some net wisdom to address. This occurred when hmisc's cut2 suddenly stopped working, compla
Hi Susan, I think that trying to do the whole thing using a graphsheet is definitely not a good idea. What others I know have done in the past is to create a poster in Powerpoint (!!!) and import the
I believe that Professor Irwin Corey (the "world's greatest expert") laid this issue to rest in the fifties: his preferred usage was "the data AM..." David O. Nelson Lawrence Livermore National Labor
Has anyone had the following problem starting up S-PLUS 2000 Release 3 on I get an error message: Error in switch(data.class(object),: invalid formula which, when traced back gives Message: Error in
Bill Dunlap, David Smith (both from Mathsoft) and I chased down this issue yesterday and achieved some sort of partial resolution to the problem. I presume that they will make some sort of official p
Any word from the people at Mathsoft regarding support for exporting to the emerging W3C Scalable Vector Graphics standard? (see http://www.w3.org/TR/SVG/ for details) David O. Nelson Lawrence Liverm
Yes, the Error() term can be a bit frustrating, but not nearly as frustrating as trying to figure out exactly what kind of analysis the C & C actually did. Once that is done, the following gives the
The SPLUS 2000 program guide says that "highly experienced users" can use the Winhelp compiler to create windows help files. Any hints as to how to get familiar with the Winhelp compiler? A search of
My thanks to the ever-informative Prof. Ripley for providing just the right pointers needed to tap into info on the windows help system. More information about Windows help files, esp. the old rtf-ba
There are at least 3 of us. I reported a similar problem in S-PLUS some months ago relating to importing into data FRAMES with missing values in the last few rows of some columns. Statsci asked me fo
Here's yet another bug-inducing "feature" added to S-PLUS 2000: A warning about importing Excel spreadsheets into S-PLUS 2000 when the "default class" is data.sheet... If the LAST ROW of the spreadsh
I seem to have uncovered a bug in the SPLUS 2000 glm support routines. (Is this the first SPLUS 2000 bug reported, and if so, do I win a prize? :-) The glm.weight() function was changed in such a way
The problem is not with as.integer, but with sqrt(), which apparently doesn't know that 5*5 = 25. Instead it computes an approximation. So, under windows v4.5: [1] -8.881784e-016 whereas under unix v
Yes, that's precisely the point I was trying to make. Perhaps it wasn't as clear as it could be because I have a tendency to mumble when I talk with my tongue in my cheek. dave David O. Nelson Lawren