[R] nonlinear regression with an iterativ function

Hi, we have a function that calculates a value for each day of the year. For example an iterative function like actvalue_0 := 0 actvalue_ {t+1} := actvalue_t + f(param_1,param_2,actvalue_t) and for some days we have measurements. Now we want to choose param_1 and param_2 so that the function

[R] simplify source code

Hi I would like to shorten mod1 - nls(ColName2 ~ ColName1, data = table, ...) mod2 - nls(ColName3 ~ ColName1, data = table, ...) mod3 - nls(ColName4 ~ ColName1, data = table, ...) ... is there something like cols = c(ColName2,ColName3,ColName4,...) for i in ... mod[i-1] - nls(ColName[i] ~

[R] append to PDF file

Hi is there a way to append a plot as PDF to an existing PDF file? savePlot seems not to have this possibility. Christof __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide

Re: [R] simplify source code

Hi Dennis, thank you very much. That works fine. Is there a possibility that R continue even if one of the models is not solvable? R currently terminates with an error message. greetings Christof Am 27-11-2011 01:34, schrieb Dennis Murphy: vars- c('y1', 'y2', 'y3') # Function to create

Re: [R] append to PDF file

Hi Jim, Hi Ken Am 27-11-2011 00:00, schrieb jim holtman: There is the 'pdftk' (PDF tool kit) that you will find on the web that will do the job. I have used it to both combine and split out the pages in the PDF file. yes, thx. GUI tools like www.pdfsam.org do the job too. But is there a

Re: [R] simplify source code

-11-2011 23:30, schrieb Christof Kluß: Hi I would like to shorten mod1 - nls(ColName2 ~ ColName1, data = table, ...) mod2 - nls(ColName3 ~ ColName1, data = table, ...) mod3 - nls(ColName4 ~ ColName1, data = table, ...) ... is there something like cols = c(ColName2,ColName3,ColName4,...) for i

[R] apply on function with vector as result

Hi, a have some code like myfunc - function(x) { ...; return c(a,b) } ys - sapply(0:100,myfunc) so I get something like c(c(a1,b1),c(a2,b2),...) But now I need the as and bs in one vector as - apply(ys, function(c(a,b)) a) bs - apply(ys, function(c(a,b)) b) Can you help me with the

Re: [R] apply on function with vector as result

Am 09-12-2011 12:54, schrieb Sarah Goslee: myfunc- function(x) {a=x; b=x-1; c(a, b) } ys- sapply(1:5, myfunc) ys [,1] [,2] [,3] [,4] [,5] [1,]12345 [2,]01234 And from there, it's not at all clear what you mean by one vector - in what order?

[R] clear plot linear mixed model

Hi, first a happy new year ;) I would like to plot the regression lines of a model like lme - lme(conc ~ name/time - 1, random=conc~time|nr,method=ML,data=measurements) Ideally a seperate plot for each regression line (respectively name) and seperate colors for the measurements (nr). How

Re: [R] clear plot linear mixed model

Am 02-01-2012 10:54, schrieb ken knoblauch: Christof Klußcklussat email.uni-kiel.de writes: lme- lme(conc ~ name/time - 1, random=conc~time|nr,method=ML,data=measurements) see plot.augPred in the nlme package thx, but how to set primary? I always get the error plot(augPred(lme))

[R] selection part of subset

Hi I want to do something like a - c(10,20,15,43,76,41,25,46) tab - data.frame(a) name - a for (v in unique(tab[[name]])) { r - subset(tab, name==v) # this does not work ... } i.e. a string on the left side of the select expression (subset). How could I solve this? thx Christof

Re: [R] selection part of subset

Hi the output should look like r - subset(tab, a==v) but now I have something like r - subset(tab, a==v) and r - subset(tab, [[a]]==v) does not work :( greetings Christof Am 05-01-2012 16:51, schrieb Christof Kluß: Hi I want to do something like a - c

Re: [R] selection part of subset

Hi thank you very much for your useful answers! In this case I solved it with Sarah's suggestion tab[tab[[name]] == v,] that works fine Greetings Christof Am 05-01-2012 16:51, schrieb Christof Kluß: Hi I want to do something like a - c(10,20,15,43,76,41,25,46) tab - data.frame

[R] RODBC vs gdata

Hi one col in my Excel file contains many numbers. But on line 3000 and some other lines are strings like FG 1. RODBS seems to omit this lines. gdata works, but is much slower. Is this a bug of RODBC or do I apply it wrong? Example with the same file.xlsx library(RODBC); excel -

Re: [R] RODBC vs gdata

:02, schrieb Christof Kluß: Hi one col in my Excel file contains many numbers. But on line 3000 and some other lines are strings like FG 1. RODBS seems to omit this lines. gdata works, but is much slower. Is this a bug of RODBC or do I apply it wrong? Example with the same file.xlsx

[R] break an axis.POSIXct

Hi I like to use axis.POSIXct to plot days from 2006 till 2008. But I only have datas for the summer months. Is it possible to get two axis breaks, to have not so long distances without points? thx Christof __ R-help@r-project.org mailing list

[R] visualize lme results

Hi for each measurement day we use lme for our four repetitions. Now it would be great if we can represent the error too. The following seems to be good. Unfortunately, there is not the full example source code. Maybe someone has done it before. (from

[R] plotting dates, incorrectly scaled x-axis?

Hi, I want to plot with axis.Date(), but something is scaled incorrectly. The red vertical line in is put on a totally wrong position. (sample below) Do you have an idea what I'm doing wrong? Thx Christof x11(width=30, height=20) x-seq(as.Date(2010-02-27), as.Date(2011-03-28),month) y -

[R] plotting dates, incorrectly scaled x-axis?

Hi, I want to plot with axis.Date(), but something is scaled incorrectly. The red vertical line in is put on a totally wrong position. (sample below) Do you have an idea what I'm doing wrong? Thx Christof x11(width=30, height=20) x-seq(as.Date(2010-02-27), as.Date(2011-03-28),month) y -

[R] linear regression with replication

Hi, we want to calculate a linear regression that results from three test series. (The Average linear regression over 3 replication.) Each test serie consists of the values ​​for 3 times (0min, 25min, 50min) and is repeated three times. We need such a regression for each combination of subject

[R] Fwd: linear regression with replication

Probably the question fits better in the lme4 mailing list, so I forwarded it. Original-Nachricht Betreff: linear regression with replication Datum: Mon, 23 May 2011 16:31:54 +0200 Von: Christof Kluß ckl...@email.uni-kiel.de An: r-h...@stat.math.ethz.ch Newsgruppen

[R] read tab delimited file from a certain line

Hi I would like to read table data from a text-files with extra informations in the header (of unknown line count). Example: informations (unknown count of lines) ... and at some point the table -- year month mday value 2013 1 16 0 ... If it was an excel

Re: [R] read tab delimited file from a certain line

(textConnection(x[i:length(x)]), header = TRUE) Hope this helps, Rui Barradas Em 16-01-2013 14:17, Christof Kluß escreveu: Hi I would like to read table data from a text-files with extra informations in the header (of unknown line count). Example: informations (unknown count of lines

Re: [R] read tab delimited file from a certain line

17, 2013 at 1:34 AM, Christof Kluß ckl...@email.uni-kiel.de wrote: Hello thank you for the fast and helpful answer! Now the following works fine for me x - readLines(filename) i - grep(^year, x) dlf - read.table(textConnection(x[i:length(x)]), header = T, stringsAsFactors=F,sep=\t

[R] lme: extract result-function

Hi, mod - lme(A ~ -1 + B+C+D+E+F+G, random = ~1 | ...) results in summary(mod)$coeff B C D E F G (Intercept) b c d e f g i Now I'm interested in the function f - function(B,C,D,E,F,G) - { return(i + b*B + c*C + d*D + e*E + f*F + g*G) } Is there a easier way to create such function with

[R] visualize correlation matrix

Hi is there a easy way to get something like http://addictedtor.free.fr/graphiques/graphcode.php?graph=137 pairs(USJudgeRatings[,c(2:3,6,1,7)], lower.panel=panel.smooth, upper.panel=panel.cor) but without the lower.panel (and without numbers and ticks at the border.) thx Christof

[R] lm without intercept, false R-squared

Hi is there a command that calculates the correct adjusted R-squared, when I work without intercept? (The R-squared from lm without intercept is false.) Greetings Chrsitof __ R-help@r-project.org mailing list

Re: [R] lm without intercept, false R-squared

run a regression I like to have the R² of the regression line and not something else. ;) Am 27-06-2012 10:25, schrieb Uwe Ligges: On 27.06.2012 09:33, Christof Kluß wrote: Hi is there a command that calculates the correct adjusted R-squared, when I work without intercept? (The R-squared

Re: [R] lm without intercept, false R-squared

run a regression I like to have the R² of the regression line and not something else. ;) Am 27-06-2012 10:25, schrieb Uwe Ligges: On 27.06.2012 09:33, Christof Kluß wrote: Hi is there a command that calculates the correct adjusted R-squared, when I work without intercept? (The R-squared

[R] save conditions in a list

Hi how would you save conditions like a = day 100; b = val 50; c = year == 2012 in a list? I like to have variables like day, val, year and a list of conditions list(a,b,c). Then I want to check if a b c is true or if a | b | c is true or similar things. Greetings Christof

Re: [R] save conditions in a list

error fun(c, year)# Should work # And now with data.frames result - lapply(conds, fun, DF) Reduce(``, result) # combine the results Reduce(`|`, result) # Hope this helps, Rui Barradas Em 02-07-2012 18:04, Christof Kluß escreveu: Hi how would you save conditions like a = day

[R] eval(parse(...)) only once in a function

Hi I would like to have something like str - df$JT == 12 fun - function(df) { b - eval(parse(str)) return(b) } but for performance eval(parse(a)) should not be evaluated at each function call, but should work as fun - function(df) { b - df$JT == 12 return(b) } Do you have an idea

Re: [R] eval(parse(...)) only once in a function

-2012 22:52, schrieb Thomas Lumley: On Mon, Sep 17, 2012 at 6:27 PM, Christof Kluß ckl...@email.uni-kiel.de wrote: but for performance eval(parse(a)) should not be evaluated at each function call, but should work as You can do it with bquote() e-parse(text=df$str==12)[[1]] e df$str == 12

[R] nlme function examples for dose-respone

Hi, I want to fit nonlinear dose-response curves, as fun(X,a,b,c), for each of our 5 trail locations. Our data basis is something like location plot year dose response For each location there are 4 plots as repetitions (over 3 years). So the interactions location*year and location*plot should

Re: [R] nlme function examples for dose-respone

. Feel free to ignore (2) and defer to the experts at (1). Cheers, Bert On Tue, Sep 25, 2012 at 8:52 AM, Christof Kluß ckl...@email.uni-kiel.de wrote: Hi, I want to fit nonlinear dose-response curves, as fun(X,a,b,c), for each of our 5 trail locations. Our data basis is something like

[R] lattice xyplot, get current level

Hi xyplot(y ~ x | subject) plots a separate graph of y against x for each level of subject. But I would like to have an own function for each level. Something like xyplot(y ~ x | subject, panel = function(x,y) { panel.xyplot(x,y) panel.curve(x,y) { #

Re: [R] lattice xyplot, get current level

subj - levels(subject) subj[panel.number()] seems to be a good solution is there something like panel.legend (instead of panel.text)? Am 02-10-2012 12:59, schrieb Christof Kluß: Hi xyplot(y ~ x | subject) plots a separate graph of y against x for each level of subject. But I would like

[R] mark sections on a time chart

Hi is there a comfortable way to mark periods on time chart (axis.Date)? To do it with arrows(...), seems to be irritating. I want to have something like ---winterspringsummer-- thx Christof __ R-help@r-project.org mailing list

Re: [R] mark sections on a time chart

Hello Rui Barradas thank you very much. That is exactly what I was looking for. Christof Am 20-10-2012 14:19, schrieb Rui Barradas: library(zoo) arrowsRange - function(from, to, at = 1, labels = NULL, length = 1/8, horizontal = TRUE, border = FALSE, ...){ require(plotrix) f1

[R] connect points in charts

Hi is there a automatic way that long distances between points are not connected. I have something like plot(x,y,type=o,...) atx - seq(as.Date(2009-04-01),as.Date(2011-04-01),month) axis.Date(1, at=atx,labels=format(atx, %b\n%Y), padj=0.5 ) but I do not want lines between points whose

[R] optimize integer function parameters

Hi I have observations obs - (11455, 11536, 11582, 11825, 11900, ...) and a simulation function f(A,B,C,D,E,F), so sim - f(A,B,C,D,E,F) e.g. sim = c(11464, 11554, 11603, 11831, 11907, ...) now I would like to fit A,B,C,D,E,F such that obs and f(A,B,C,D,E,F) match as well as possible. A,..,F

Re: [R] optimize integer function parameters

- rmse } } Am 23-07-2013 13:20, schrieb Enrico Schumann: On Tue, 23 Jul 2013, Christof Kluß ckl...@email.uni-kiel.de writes: I have observations obs - (11455, 11536, 11582, 11825, 11900, ...) and a simulation function f(A,B,C,D,E,F), so sim - f(A,B,C,D,E,F) e.g

[R] alternative to rbind for data.table

Hi I want to add a row to a data.table in each round of a for loop. rbind seems to be a inefficient way to implement this. How would you do this? The slow solution: library(data.table) Rprof(test.out) dt - data.table() for (i in (1:1)) { # algorithm that generates a list with different

[R] self-starter functions for y = a + b * c^x

Hi there are some predefined self-start functions, like SSmicmen, SSbiexp, SSasymp, SSasympOff, SSasympOrig, SSgompertz, SSflp, SSlogis, SSweibull, Quadratic, Qubic, SSexp (nlrwr) Btw, do you know graphic examples for this functions? The SSexpDecay (exponential decay) for y = (y0 -

Re: [R] self-starter functions for y = a + b * c^x

Hi Ken Am 14-08-2012 11:50, schrieb ken knoblauch: Christof Kluß ckluss at email.uni-kiel.de writes: But now I search a self-starter for y = a + b * c^x You might be able to adapt the SS.calib function from the psyphy package for this. It was designed to fit gamma functions

Re: [R] self-starter functions for y = a + b * c^x

Am 14-08-2012 19:40, schrieb Peter Ehlers: On 2012-08-14 00:09, Christof Kluß wrote: But now I search a self-starter for y = a + b * c^x Can't you just reparameterize with c^x = exp(x * log(c)) ? oh, I've missed this... thank you very much! Greetings Christof

[R] force pdf()

Hi is there a way to generate a pdf, e.g. pdf(test.pdf), even if the file test.pdf is open? (e.g. with acrobat) i.e. I would have to close test.pdf in the viewer without any user interaction, than override the file and maybe open it again. thx Christof

[R] make R program faster

Hi there are some good tips in The R Inferno http://www.burns-stat.com/documents/books/the-r-inferno/ or connect C++ to R with Rcpp http://dirk.eddelbuettel.com/code/rcpp.html or byte code compiler (library(compiler)) or library(data.table) but do you have an idea to fasten standard R source