similar to: Self-starting nls functions

Hello, nls library provides 6 self-starting models, among them: SSfp, a four parameters logistic function. Its self-starting procedure involves several steps. One of these steps is: pars <- as.vector(coef(nls(y ~ cbind(1, 1/(1 + exp((xmid - x)/exp(lscal)))), data = xydata, start = list(lscal = 0), algorithm = "plinear"))) which assumes an initial value of lscal equal to 0. If lscal

>>>>> "PG" == Philippe Grosjean <phgrosje at ulb.ac.be> writes: PG> I am interested by your list of students wishes, since I am PG> developing a GPL user interface for math/stat calculation PG> engines (among others: R and Splus, but also Octave, Scilab PG> and Mathematica,...). It will run under Windows only (for the PG> moment),

Hi, I have 16 tables of 39 variables with around 1500 to 3500 cases in each table. These 16 tables are in 16 different data frames. I want to merge them in a single large data frame, so: All <- rbind(T1, T2, ...) The problem: it is VERY slow, and takes a lot of memory (> 500 Mb) on my Athlon XP1800+, 1.75 Go memory, Win 2000 Pro SP1, R 1.4.1. Any suggestion to improve speed, and/or memory

Hi, I have submitted a package to CRAN in ftp://ftp.tuwien.ac.at/incoming/pastecs_1.0.tar.gz. The R extension manual advises to send a mail to cran at ci.tuwien.ac.at about it, but it fails. I have also tried WWWadmin at ci.tuwien.ac.at. Are these email adresses still valid? Best, Philippe Grosjean ...........]<(({?<...............<?}))><............................... ) ) ) )

It's two days I try to answer to a mail of Achim Zeileis, the maintainer of strucchange package, using the mail address: zeileis at ci.tuwien.ac.at. I got the following error: ----- The following addresses had permanent fatal errors ----- <zeileis at ci.tuwien.ac.at> (reason: 550 5.7.1 <zeileis at ci.tuwien.ac.at>... Access denied) Can anyone tell me how to reach him?

using: identify(...., plot=FALSE) allows identifying points in a graph without printing the corresponding label in the graph. This works fine in R for Windows. However, when I copy the graph as metafile into Word, the corresponding label appears in black font in the copied picture. Obviously, identify() prints the label anyway, in "invisible" font or so, in the graph, and this

Generalised Inverse: The Moore-Penrose Generalisied Inverse is probably better defined as a pseudo-Inverse that arises in solving least squares problems. Another well known pseudo-Inverse is the so-called Drazin pseudo-Inverse. If memory serves (and it's been 10-12 years!) it can be obtained via a diagonalisation. Anyway, I dare say Prof. Ripley (among others) probably has "all the

Hello, I would like to plot two superposed surfaces (original data, and smoothed models). Basically, I would like to create a usual persp plot with box, and then superpose a second surface (without box and axes) on the top plane of the box defined by the first graph. I don't know if it is clear :-( Is it possible to draw such a second surface in an existing persp plot in R? Another (related)

Hello all, Since I got no answer to my first mail, I suspect I did not formulate my question adequately, so I try again. In R, one can debug or trace a function 'foo' with debug(foo) or trace(foo), respectively. This leads to a special treatment of the function 'foo' until one enter undebug(foo) or untrace(foo). I would like to know if there is a convenient way to know at any

Hi, Under Windows, I am exporting graphs from R (in metafile -emf- format), and want to combine them using a graphic package. It happens that the background of the graph is represented by a filled rectangle (with the color given by bg=...). This is to be expected. However, I find no options to disable this and produce a graph with transparent background. Does such an option exist? If not, then

Hello everybody, I am making a modified version of "Stephan Steinhaus' benchmark test for number crunching, v. 2, (see http://www.scinetificweb.com/ncrunch/ncrunch.pdf for the original version), comparing several functions of some math/stat software. R is not performing bad at all... except for the sorting of a 1,100,000 random vector (test #3) which is the worst of all (see cell F3 in

Hello everybody, I am making a modified version of "Stephan Steinhaus' benchmark test for number crunching, v. 2, (see http://www.scinetificweb.com/ncrunch/ncrunch.pdf for the original version), comparing several functions of some math/stat software. R is not performing bad at all... except for the sorting of a 1,100,000 random vector (test #3) which is the worst of all (see cell F3 in

Thorsten is right. There is a direct formula for computing the Moore-Penrose inverse using the singular value composition of a matrix. This is incorporated in the following: mpinv <- function(A, eps = 1e-13) { s <- svd(A) e <- s$d e[e > eps] <- 1/e[e > eps] return(s$v %*% diag(e) %*% t(s$u)) } Hope it helps. Dietrich

Hi R help, I am trying to determine how nls() generates a function based on the self-starting SSlogis and what the formula for the function would be. I've scoured the help site, and other literature to try and figure this out but I still am unsure if I am correct in what I am coming up with. ************************************************************************** dat <-

Hello, I am trying to fit a nlme (non linear mixed effect). I am using the SelfStart function SSlogis. However the data in my hand contains few observations per subject (4 or less), so the nlsList doesn't work... In this case I should fixe initial parameter estimates. I remark that values of initial estimates have a greater effect on the model fit (i.e. loglikelihood, AIC and also on

I'm wondering whether AIC scores extracted from nls() objects using AIC() are based on the correct number of estimated parameters. Using the example under nls() documentation: > data( DNase ) > DNase1 <- DNase[ DNase$Run == 1, ] > ## using a selfStart model > fm1DNase1 <- nls( density ~ SSlogis( log(conc), Asym, xmid, scal ), DNase1 ) Using AIC() function: >

Hi I want to change a control parameter for an nls () as I am getting an error message "step factor 0.000488281 reduced below 'minFactor' of 0.000976562". Despite all tries, it seems that the control parameter of the nls, does not seem to get handed down to the function itself, or the error message is using a different one. Below system info and an example highlighting the

Hi folks, I am tearing my hair out on this one. I am using an example from Pinheiro and Bates. ### this works data(Orange) mod.lis <- nlsList(circumference ~ SSlogis(age, Asymp, xmid, scal), data=Orange ) ### This works mod <- nlme(circumference ~ SSlogis(age, Asymp, xmid, scal), data=Orange, fixed = Asymp + xmid + scal ~ 1, start =

Full_Name: Hendrik Weisser Version: 2.6.1 OS: Linux Submission from: (NULL) (139.19.102.218) The following computation never finishes and locks R up: > values <- list(x=10:30, y=c(23.85, 28.805, 28.195, 26.23, 25.005, 20.475, 17.33, 14.97, 11.765, 8.857, 5.3725, 5.16, 4.2105, 2.929, 2.174, 1.25, 1.0255, 0.612, 0.556, 0.4025, 0.173)) > y.max <- max(values$y) > model <- nls(y ~

Dear all, I am trying to fit a non linear regression model to time series data. If I do this: reg.logis = nls(myVar~SSlogis(myTime,Asym,xmid,scal)) I get this error message (translated to English from French): Erreur in nls(y ~ 1/(1 + exp((xmid - x)/scal)), data = xy, start = list(xmid = aux[1], : le pas 0.000488281 became inferior to 'minFactor' of 0.000976562 I then tried to set