similar to: lm-residuals and NA {was Can't understand error message :-{}

There needs to be more recognition in the academic community of the importance of contributions to widely used software. While many organizations are encouraging software development if it generates revenue, it is harder to measure the value of software that you give away. It is often hard to measure the number of users, so unless they mention you in citations, there is little hard evidence to

I'm trying to extract the points above and below a particular lsfit. I can only get the residuals from the original fit though. x = runif(100, 0, 10) plot(x) abline(lsfit(1:100, test)) abline(lsfit(1:100, test + sd(test))) #I want the points above THIS line. Is there a way to use the coefficients from the fit to do this? Thanks for any help.

I'm trying to understand how to use kpss.test() properly. If I have a level stationary series like rnorm() in the help page, shouldn't I get a small p-value with the null hypothesis set to "Trend"? The (condensed) output from kpss.test() for the two possible null hypotheses is given below. I don't see any significant difference between these results. > x <-

Bill Venables <wvenable at attunga.stats.adelaide.edu.au> wrote: [ snip ] > I don't say this will appeal to everyone straight away, but I > would even favour using a binary operator syntax for the > assignment function form of the above replacement function, for > example > > "%cast.as%" <- function(x, form) { > storage.mode(x) <-

Ross Ihaka <ihaka at stat.auckland.ac.nz> wrote: > Martin Maechler writes: [ snip ] > > I've seen > > As.numeric <- function(x) { storage.mode(x) <- "numeric"; x } > > solving the above problem. > > > > What do you (R-help readers !) think? > > I like the idea, but maybe not the name so much (too much scope for >

John Logsdon <j.logsdon at lancaster.ac.uk> wrote: > This is a problem that has been bothering me for some time. At the > moment, I use RedHat Linux, Octave, occasionally R and other FSF/GPL > products. They are generally better, at least in their fundamental > design, than the commercial versions but there is no clear mechanism to > ensure their continued development.

Ross Ihaka <ihaka at stat.auckland.ac.nz> wrote: [ in response to discussions about S compatibility ] > Some things we almost certainly won't be changing. Scoping is one of > these (although I have to confess that I have made the pitch to Robert > that we should drop our scoping in favour of the S version - he > sensibly said "no"). I have a question/comment on

Hi, I'm trying to install rimage on a Mac OS X 10.4 machine. I followed the advice in previous R-help threads and got over the hurdles of having the header files in the right places, among other things. But I can't figure out what to do with this error. ice.pnl.gov:/home/waichler<949>system_profiler -detailLevel mini SPSoftwareDataType Software: System Software Overview:

I frequently use Makefiles to manage routine data manipulations and repeated analyses where the "raw" data tend to change with time. With S/Splus, this has worked quite naturally, as long as I'm careful to use S object names that map directly to Unix/Linux filenames. For example, say I'm building a particular graphic (chart.ps) from a data.frame in S (mydata) that comes from an

Peter Dalgaard BSA <p.dalgaard at biostat.ku.dk> wrote: > Martin Maechler <maechler at stat.math.ethz.ch> writes: [ snip... ] > > >From this logic, coercing to a matrix should coerce to character rather > > than integer/numeric codes. > > Hm. I'm not sure there really is much logic in this... Factors are > sometimes character-like, sometimes

Sorry for the off-topic (non-R) post. Has anyone seen/tried this (from this week's NA-digest)? Andy ------------------------------------------------------- From: Joel Malard <JM.Malard at pnl.gov> Date: Sat, 01 May 2004 15:31:15 -0700 Subject: ACRE, Parallel Covariance Component Estimation Code A couple of people have asked recently for a copy of the parallel (restricted/residual)

In the following special case, seq fails to give the right answer (which is 0 ) > seq(0,0,1) Error in if (dd < sqrt(.Machine$double.eps)) return(from) : missing value where logical needed For any other equal from and to , it works: > seq(1,1,1) [1] 1 The error occurs in the statement dd <- abs(del)/max(abs(to), abs(from)) of seq.default -- for obvious reasons.

How can I access examples of trellis graphs in R? I'm especially interested in getting code for xyplot() examples. Failing a collection of examples in R, I see that under Splus there are numerous example.*() functions. I can access Splus with a command line but can't open a graphics window, so I can't to execute the example.*() functions. Is there a way I could get the associated

>>>>> "WSt" == Werner Stahel <stahel@stat.math.ethz.ch> writes: [to me privately] WSt> I do not want to call this a bug, but it bugged me. Please try the WSt> following: WSt> tform <- sqrt(RADAI) ~ sqrt(RADAI.e) + TAGE.ej + SPITAL + ARZT + DAS28 + WSt> SJC + TJC + DGA + HAQ + PGA + PAIN + YEAR.SYM + YEAR.DIA + WSt> ALTER +

I have just fitted a lm model containing a bs smooth term. I asked Martin how I could see the estimated function. He said that such a possibility -- which is available in S-Plus' plot.gam -- was on your list of desirable functions. This is a vote to give it high priority. (I appreciate all your efforts!) Werner Stahel ------------------ This message was sent by

Hi, [I'mm cc-ing r-help on this, since some others might be interested (Tony Rossini is, at least)] Yes, I saw your mail and was wondering about what would be the best approach to do it. It's definitely not trivial, mainly because Trellis is not designed to handle factors on the x-axis (chiefly for convenience, as factors typically have long names which are difficult to write on the

Dear R-devel, It seems that lsfit incorrectly reports coefficients when the input matrix 'x' is rank-deficient, see the example below: ## here values of 'b' and 'c' are incorrectly swapped > x <- cbind(a=rnorm(100), b=0, c=rnorm(100)); y <- rnorm(100); lsfit(x, y)$coef Intercept a b c -0.0227787 0.1042860 -0.1729261 0.0000000 Warning

Can anyone tell me how to obtain a custom scale for the colorkey in each levelplot of a lattice? I am using lattice and levelplot to plot z = f(x, y) for multiple z variables. x and y values are the same across plots, but units for z are different and therefore I need to create a custom scale for each plot's key. It follows that I need to place each plot's key inside that plot's

When I generate levelplot figures with Postscript, the lines from the underlying grid (i.e., the x and y in z ~ x* y) show up as faint white lines in the output. If I have a dense grid, the lines wash out the contour colors. I can avoid the problem if I use png instead, but that brings up other problems. Any ideas for how I can get Postscript to not show those lines around the cells? Scott

I wanted a simple function for bilinear interpolation on a 2-D grid, and interp.surface() in the fields package didn't quite suit my needs. In particular, it requires uniform spacing between grid points. It also didn't have the "visual" reference frame I was looking for. Here is an alternative function, followed by an example. # A function for bilinear interpolation on a 2-d