Re: [R] Reshape

HI, A minor change to make it a bit more generic.  res <- reshape(ddply(df,.(Person), mutate,id=((seq_along(Person)-1)%%2+1)),idvar=c("Person","Date"),timevar="id",direction="wide")   colnames(res)[3:4] <- paste0("Amt",1:2)  rownames(res) <- 1:nrow(res) In cases like: df1 <- rbind(df,data.frame(

Re: [R] Reshape

Hi, Try: library(plyr)  res <- reshape(ddply(df,.(Person),mutate,id=rep(1:2,length(Person)/2)),idvar=c("Person","Date"),timevar="id",direction="wide")  colnames(res)[3:4] <- paste0("Amt",1:2) rownames(res) <- 1:nrow(res)   res #  Person   Date Amt1 Amt2 #1  1 12/01/2012  100   10 #2 

Re: [R] Data handling

Thanks A.K. and Jim! Thanks very much, both solution works fine! But I can´t figure out what was the problem with my code. Make step-by-step is not recommended? Is just that difference? Thanks again for help! Raoni 2013/10/15 arun > Try: > op <- options(digits.secs=4) > > TimeCC <- as.POS

[R] saveXML() prefix argument

I'm using the "XML" package and specifically the saveXML() function but I can't get the "prefix" argument of saveXML() to work: library("XML") concepts <- c("one", "two", "three") info <- c("info one", "info two", "info three") root <- newXMLNode("root") for (i in 1:length(concepts)) { cu

[R] Reshape

Hello, I have data frame like the one created below, ##input data Person <- c(1,1,1,1,1,1,2,2) Amount <- c(100,10,100,10,100,10,50,150) Date <- c("12/01/2012", "12/01/2012", "01/01/2013","01/01/2013","02/01/2013","02/01/2013","12/01/2012","12/01/2012") df <- data.frame(Person,Amount,Date)

Re: [R] Error message after following Appendix A R-intro

Hello, The command you have to use is "lm", not "1m". Regards, Pascal On 17 October 2013 08:03, timscharlton wrote: > I am new to R and tried to go through the commands shown at App. A of > "R-intro" (p.78). > > I get the following error message. > >> fm<-1m(y~x,data=dummy) > Error: unexpected

Re: [R] Error message after following Appendix A R-intro

On Oct 16, 2013, at 4:03 PM, timscharlton wrote: > I am new to R and tried to go through the commands shown at App. A of > "R-intro" (p.78). > > I get the following error message. > >> fm<-1m(y~x,data=dummy) > Error: unexpected symbol in "fm<-1m" In some screen fonts it's difficult to tell t

Re: [R] Error message after following Appendix A R-intro

Hi, On Oct 16, 2013, at 7:03 PM, timscharlton wrote: > I am new to R and tried to go through the commands shown at App. A of > "R-intro" (p.78). > > I get the following error message. > >> fm<-1m(y~x,data=dummy) > Error: unexpected symbol in "fm<-1m" I think the example shows the function

Re: [R] using ddply with segmented regression

Hi Paul, Thanks for starting this thread. I have been struggling with the same plyr problems. I have also had some trouble moving onto the next step - integrating the segmented output with ggplot2 or lattice. Have you had any luck in this? There is good documentation on adding lm into ggpl

[R] Error message after following Appendix A R-intro

I am new to R and tried to go through the commands shown at App. A of "R-intro" (p.78). I get the following error message. > fm<-1m(y~x,data=dummy) Error: unexpected symbol in "fm<-1m" I suspect it might be related to the tilda. All the commands in R-intro for the tilda show it in a superscri

Re: [R] How to obtain restricted estimates from coxph()?

Thanks very much for your help, Terry and Göran! As pointed out by Göran, the difficult part is that it's an open set. How to obtain a valid MLE in this case? Thanks, YH On Wed, Oct 16, 2013 at 9:55 AM, Göran Broström wrote: > > > On 2013-10-16 14:33, Terry Therneau wrote: > >> >> >> On

Re: [R] Plot time series data irregularly hourly-spaced

You could bump up the day each time an hour was less than the previous one. E.g., testtime <- c("20:00:00","22:10:00","22:20:00","23:15:00","23:43:00","00:00:00","00:51:00","01:00:00") var <- seq_along(testtime) # so you know what the plot should look like # turn it ino a POSIXlt object so

Re: [R] Plot time series data irregularly hourly-spaced

You just need the date, otherwise how would it know what time comes first? In strptime(), a date is being assumed. Try this: testtime<-c("20:00:00","22:10:00","22:20:00","23:15:00","23:43:00","00:00:00","00:51:00","01:00:00") testday <- rep(Sys.Date() - c(1,0), times = c(5,3)) plot(as.POSIXct(

Re: [R] Plot time series data irregularly hourly-spaced

Hi, This may get you started. testtime1 <- factor(testtime,levels=testtime)  plot(as.numeric(testtime1),var,type="b",xlab="Time",ylab="Var",xaxt="n")  axis(1,at= as.numeric(testtime1), labels=levels(testtime1)) ## labels are not spaced according to time interval #Another idea would be: testti

Re: [R] Small p from binomial probability function.

On 10/16/13 23:56, Benjamin Ward (ENV) wrote: Hi, Thanks again for your answers, just so as I can get clear what is happening, with the uniroot method, I'm defining a function in which the binomial probability function pbinom is present but in addition p0 is subtracted from the result - in th

[R] Plot time series data irregularly hourly-spaced

Dear all, I have a time series of data that I would like to represent in a plot. But I am facing some problems to do it because the time is represented in "hours", it can start in one day and end in another day, and it is not regularly spaced. My problem is that when I plot my data, my X-axis alw

[R] Workshop: structural equation modeling in R

Dear r-help list members, I'm teaching a one-day workshop on "An Introduction to Structural Equation Modeling with the sem Package for R" at McMaster University in Hamilton, Ontario, Canada, on November 22. The workshop is open to non-McMaster attendees at a small charge. Further information, incl

Re: [R] Problem with lapply

Ah ok, I see the problem! Thank you again! -- View this message in context: http://r.789695.n4.nabble.com/Problem-with-lapply-tp4678290p4678371.html Sent from the R help mailing list archive at Nabble.com. __ R-help@r-project.org mailing list https:/

Re: [R] identifying which column an observation comes from?

In cases like these: mat <- matrix(data = c(1,0,0,0,1,0,0,0,1,1,1,0), nrow = 3, byrow = T)   rownames(mat) <- c("AL", "MS", "FL") rowSums(col(mat) * mat) #AL MS FL # 1  1  6 A.K. On Wednesday, October 16, 2013 4:18 PM, Bert Gunter wrote: ?col rowSums(col(mat) * mat) produces a named v

Re: [R] identifying which column an observation comes from?

?col rowSums(col(mat) * mat) produces a named vector that you can convert to a data frame if you like, although it's really not necessary. Cheers, Bert On Wed, Oct 16, 2013 at 12:49 PM, arun wrote: > Hi, > Try: > > ind <- which(mat==1,arr.ind=TRUE)[,2] > dat1<- data.frame(Code=names(ind),ind

Re: [R] identifying which column an observation comes from?

Hi, Try: ind <- which(mat==1,arr.ind=TRUE)[,2] dat1<- data.frame(Code=names(ind),ind=ind,stringsAsFactors=FALSE)  row.names(dat1) <- 1:nrow(dat1) A.K. On Wednesday, October 16, 2013 3:29 PM, Karl Fetter wrote: Hello, I have a matrix of samples (rows) and haplotypes (columns), where 0 indic

Re: [R] Ylim problem - plot.correlog, ncf package

Hi Pyh- having the same problem- did you ever find a solution? Cheers, Ben On Saturday, August 11, 2012 8:41:21 AM UTC-7, Pyhrell wrote: > > Hi, > > I'm doing cross-correlation correlograms with the ncf package. I have four > study sites ; four correlograms. > I'd like to get the same y scale

[R] identifying which column an observation comes from?

Hello, I have a matrix of samples (rows) and haplotypes (columns), where 0 indicates that a sample does not posses that columns haplotype and 1 indicates it does. So sample1 has 0's for every column, except the column that represents haplotype X, and it has a 1. I want a length(sample) x 2 dataf

Re: [R] problem with MLE estimation using Kalman filter

Thanks Duncan and Ingmar, everything is all good now. Best On Wed, Oct 16, 2013 at 3:12 PM, Duncan Murdoch wrote: > On 16/10/2013 2:57 PM, Curtis Burkhalter wrote: > >> Thanks, that always gets me for some reason. Now when I run it though I >> get an error message at the very end that states "

Re: [R] problem with MLE estimation using Kalman filter

On 16/10/2013 2:57 PM, Curtis Burkhalter wrote: Thanks, that always gets me for some reason. Now when I run it though I get an error message at the very end that states "could not find function "M.n". I don't understand why I'm getting this message b/c there is no where that calls a function nam

Re: [R] Assign date according to defined time interval

Hi Weijia, This will give you the rownames of the split variables. lst1 <- split(a,list(a$COUNTRY,a$SITEID))  res <- t(sapply(lst1,function(x) {                 x$SCRNDT <- as.Date(x$SCRNDT, "%d-%b-%y")                 unlist(lapply(split(b,b$TMPT),function(y){                   sum(x$SCRNDT >=

Re: [R] problem with MLE estimation using Kalman filter

Thanks, that always gets me for some reason. Now when I run it though I get an error message at the very end that states "could not find function "M.n". I don't understand why I'm getting this message b/c there is no where that calls a function named "M.n" and I don't define a function with that n

Re: [R] problem with MLE estimation using Kalman filter

On Wed, Oct 16, 2013 at 8:30 PM, Curtis Burkhalter wrote: > I try to use the mle2 function written for R. The error message states > that the argument "minuslog1" is missing with no default, but I've The argument is minuslogl Note: l instead of 1 hth, Ingmar _

[R] problem with MLE estimation using Kalman filter

Hello, I've recently run into a problem when trying to do some MLE parameter estimation in R using a non-linear Kalman filter. I've written some of the functions myself and everything seems to be working fine until the end when I try to use the mle2 function written for R. The error message stat

Re: [R] How to write an error to output

Dear sir, Thanks a lot for your wonderful suggestion. Regards Katherine On Wednesday, 16 October 2013 5:28 PM, jim holtman wrote: Will this work for you: mydat = data.frame(A = c(19, 20, 19, 19, 19, 18, 16, 18, 19, 20), B = c(19, 20, 20, 19, 20, 18, 19, 18, 17, 16)) if (length(mydat$A)

[R] Extract a predictors form constparty object (CHAID output) in R

I have a large dataset (questionnaire results) of mostly categorical variables. I have tested for dependency between the variables using chi-square test. There are an incomprehensible number of dependencies. I used the chaid() function in the CHAID package to detect interactions and separate out (w

Re: [R] map with inset

I've had difficulty getting subplot() to work with maps (the TeachingDemos and Hmisc versions seem to be the same). This will work, but there should be a better way. You have to print the inset map first or you will get an error message. require(maps) require(mapdata) Layout <- matrix(c(rep(2, 3),

Re: [R] new dprep package for windows

On Oct 16, 2013, at 8:17 AM, peter dalgaard wrote: > Look at > > http://cran.r-project.org/web/packages/dprep/index.html > > and follow the link. You'll find that > > (a) There is a newer version > (b) CRAN no longer carries binaries for this package, presumably because it > won't build and t

Re: [R] Assign date according to defined time interval

HI Weijia, Please check whether this is what you wanted. Weijia <- load("/home/arunksa111/Downloads/arun_help.RData" ) a[sapply(a,is.factor)] <-lapply(a[sapply(a,is.factor)],as.character) str(a)  b[sapply(b,is.factor)] <- lapply(b[sapply(b,is.factor)],as.character) str(b)  b$DT_ETP <- as.Date(b$D

Re: [R] map with inset

... and if that doesn't do it, check out the CRAN "spatial" task view, The "maptools" package or others might have what you need with a friendlier interface. Cheers, Bert On Wed, Oct 16, 2013 at 9:39 AM, David Winsemius wrote: > > On Oct 16, 2013, at 3:47 AM, markw wrote: > > > Hi, > > > > I a

Re: [R] map with inset

On Oct 16, 2013, at 3:47 AM, markw wrote: > Hi, > > I am trying to put an inset of North America onto a finer-scale map and > cannot seem to get the two maps on the same plot. > > the main map is: > map("worldHires", c("Canada", "USA"), xlim=c(-75, -52), ylim=c(40, 55), > col="gray90", fill=TR

Re: [R] Math notation in a title

On Oct 15, 2013, at 10:52 AM, David Arnold wrote: > Hi, > > I'd like to put the following in a main title of a plot. > > > > Can someone show me how to do this? ?plotmath plot(1, main=expression("Pareto Distribution:"~alpha==2*","~ital

Re: [R] Is there something wrong with R version 3.0.2 (2013-09-25) -- "Frisbee Sailing"?

Tom, Under that same version on Windows 7, I receive the following: > pnorm(-1.53,0,1) [1] 0.06300836 R version 3.0.2 (2013-09-25) -- "Frisbee Sailing" Copyright (C) 2013 The R Foundation for Statistical Computing Platform: x86_64-w64-mingw32/x64 (64-bit) With gratitude, CEO'Riley Jr. Charl

Re: [R] How to obtain restricted estimates from coxph()?

On 2013-10-16 14:33, Terry Therneau wrote: On 10/16/2013 05:00 AM, r-help-requ...@r-project.org wrote: Hello, I'm trying to use coxph() function to fit a very simple Cox proportional hazards regression model (only one covariate) but the parameter space is restricted to an open set (0, 1). C

Re: [R] Is there something wrong with R version 3.0.2 (2013-09-25) -- "Frisbee Sailing"?

I cannot confirm your report. I get 0.63 under 3.0.2 (2013-09-25), x86_64-pc-linux-gnu (64bit), and x686-pc-linux-gnu (32bit). --- Jeff NewmillerThe . . Go Live... DCN:Basics

Re: [R] new dprep package for windows

Look at http://cran.r-project.org/web/packages/dprep/index.html and follow the link. You'll find that (a) There is a newer version (b) CRAN no longer carries binaries for this package, presumably because it won't build and the maintainer isn't maintaining it. So, if you want it enough, you nee

Re: [R] Is there something wrong with R version 3.0.2 (2013-09-25) -- "Frisbee Sailing"?

On Wed, 16 Oct 2013, tom soyer wrote: pnorm(-1.53,0,1) under version 3.0.2 gives 0.05155075. I am pretty sure it should be 0.063. Is there something wrong with this version of R? Tom, Running 3.0.2 here on Slackware: pnorm(-1.53,0,1) [1] 0.06300836 Rich -- Richard B. Shepard, Ph.D.

Re: [R] Is there something wrong with R version 3.0.2 (2013-09-25) -- "Frisbee Sailing"?

On 16-Oct-2013 14:54:00 tom soyer wrote: > Hi, > > pnorm(-1.53,0,1) under version 3.0.2 gives 0.05155075. I am pretty sure it > should be 0.063. Is there something wrong with this version of R? > > I am using: > R version 3.0.2 (2013-09-25) -- "Frisbee Sailing" > Copyright (C) 2013 The R Foundati

Re: [R] Cleaning up workspace

This has been reported before on the bug list (https://bugs.r-project.org/bugzilla3/show_bug.cgi?id=15481). The message is coming from the methods package, but I don't know if it's a bug or ignorable. Duncan Murdoch On 16/10/2013 11:03 AM, Prof J C Nash (U30A) wrote: In order to have a clea

Re: [R] Is there something wrong with R version 3.0.2 (2013-09-25) -- "Frisbee Sailing"?

Ok. Thanks Peter. It was my bad - typo. You caught it. Sorry everyone. On Wed, Oct 16, 2013 at 10:03 AM, peter dalgaard wrote: > > On Oct 16, 2013, at 16:54 , tom soyer wrote: > > > Hi, > > > > pnorm(-1.53,0,1) under version 3.0.2 gives 0.05155075. I am pretty sure > it > > should be 0.063. Is

Re: [R] Is there something wrong with R version 3.0.2 (2013-09-25) -- "Frisbee Sailing"?

On 16/10/2013 10:54 AM, tom soyer wrote: Hi, pnorm(-1.53,0,1) under version 3.0.2 gives 0.05155075. I am pretty sure it should be 0.063. Is there something wrong with this version of R? I am using: R version 3.0.2 (2013-09-25) -- "Frisbee Sailing" Copyright (C) 2013 The R Foundation for Statist

Re: [R] new dprep package for windows

Contact the maintainer? ?maintainer If that doesn't work (it doesn't seem to have worked for the CRAN volunteers), then download the source code from the archives and fix it? This may require you to learn more than you expected to but in some cases may be the only option. -

Re: [R] Is there something wrong with R version 3.0.2 (2013-09-25) -- "Frisbee Sailing"?

On both 32 and 64 bit Windows I get .063. -- Bert On Wed, Oct 16, 2013 at 7:54 AM, tom soyer wrote: > Hi, > > pnorm(-1.53,0,1) under version 3.0.2 gives 0.05155075. I am pretty sure it > should be 0.063. Is there something wrong with this version of R? > > I am using: > R version 3.0.2 (2013-

Re: [R] Is there something wrong with R version 3.0.2 (2013-09-25) -- "Frisbee Sailing"?

On Oct 16, 2013, at 16:54 , tom soyer wrote: > Hi, > > pnorm(-1.53,0,1) under version 3.0.2 gives 0.05155075. I am pretty sure it > should be 0.063. Is there something wrong with this version of R? Not on my system (Mac OS X): > pnorm(-1.53,0,1) [1] 0.06300836 But check your typing, and/or ey

[R] Cleaning up workspace

In order to have a clean workspace at the start of each chapter of a book I'm "knit"ing I've written a little script as follows: # chapclean.R # This cleans up the R workspace ilist<-c(".GlobalEnv", "package:stats", "package:graphics", "package:grDevices", "package:utils", "package:datasets",

[R] Is there something wrong with R version 3.0.2 (2013-09-25) -- "Frisbee Sailing"?

Hi, pnorm(-1.53,0,1) under version 3.0.2 gives 0.05155075. I am pretty sure it should be 0.063. Is there something wrong with this version of R? I am using: R version 3.0.2 (2013-09-25) -- "Frisbee Sailing" Copyright (C) 2013 The R Foundation for Statistical Computing Platform: i686-pc-linux-gnu

[R] new dprep package for windows

Hi experts, I need the new version of dprep package for windows. What I have found is version 1. if I am installing version 1 I get this error *Error: package ‘dprep’ was built before R 3.0.0: please re-install it* * * Is ther a new version of this package? if not what should I do? [[

Re: [R] how to merge GRange object?

On 10/16/2013 06:32 AM, John linux-user wrote: Hello everyone, I am wondering how to simply merge two GRanges objects by range field and add the value by additional vector. For example, I have two objects below Hi -- GRanges is from a Bioconductor package, so please ask on the Bioconductor

[R] how to merge GRange object?

Hello everyone, I am wondering how to simply merge two GRanges objects by range field and add the value by additional vector. For example, I have two objects below    obj1  seqnames           ranges strand |       Val                       |   [1] chr1_random [272531, 272571]      + |        8

[R] map with inset

Hi, I am trying to put an inset of North America onto a finer-scale map and cannot seem to get the two maps on the same plot. the main map is: map("worldHires", c("Canada", "USA"), xlim=c(-75, -52), ylim=c(40, 55), col="gray90", fill=TRUE) map.axes() map('rivers', add=TRUE) map.scale(-73, 54, re

Re: [R] plot: want only dots

Thanks. Yes it was due to the factor-thing. If I change Genotype as mentioned everything works fine. Thanks Hermann 2013/10/15 Sarah Goslee > Hi, > > Genotype is a factor, and R is giving you the default type for that data > type. > > I changed your data frame to mydata because df() is a functi

Re: [R] Small p from binomial probability function.

Hi, Thanks again for your answers, just so as I can get clear what is happening, with the uniroot method, I'm defining a function in which the binomial probability function pbinom is present but in addition p0 is subtracted from the result - in this case p0 is the large P I want to plug in so 0

[R] How to obtain restricted estimates from coxph()?

Hello, I'm trying to use coxph() function to fit a very simple Cox proportional hazards regression model (only one covariate) but the parameter space is restricted to an open set (0, 1). Does anyone know how to obtain an estimate in a particular parameter space by using coxph function? Or any othe

Re: [R] How to obtain restricted estimates from coxph()?

On 10/16/2013 05:00 AM, r-help-requ...@r-project.org wrote: Hello, I'm trying to use coxph() function to fit a very simple Cox proportional hazards regression model (only one covariate) but the parameter space is restricted to an open set (0, 1). Can I still obtain a valid estimate by using co

Re: [R] How to write an error to output

Will this work for you: mydat = data.frame(A = c(19, 20, 19, 19, 19, 18, 16, 18, 19, 20), B = c(19, 20, 20, 19, 20, 18, 19, 18, 17, 16)) if (length(mydat$A) > 10) { write.csv(data.frame(error = "A has length more than 10"), 'result.csv', row.names = FALSE) stop("A has length more than 10") }els

[R] How to write an error to output

Dear R forum, The example below is just an indicative one and I have constructed it. My real life data and conditions are different. I have a data.frame as given below mydat = data.frame(A = c(19, 20, 19, 19, 19, 18, 16, 18, 19, 20), B = c(19, 20, 20, 19, 20, 18, 19, 18, 17, 16)) if (length(m

[R] R reference group in Cape Town South Africa

Hi R users in Cape Town, I will be starting the Epidemiology/Clinical Research program at the University of Cape Town in the Summer of Jan 2014 and so I am looking to meet any useRs. I assume there is a refeRence group in Cape Town? Please do email me as I would love to corRespond with you, Best

Re: [R] Finding solution for non-linear equations

On Oct 15, 2013, at 17:18 , Berend Hasselman wrote: > > On 15-10-2013, at 15:24, Ron Michael wrote: > >> Hi, >> >> I need to solve following simultaneous equations for A, B, Y1, Y2: >> >> B * Phi(Y1 - A) + (1-B) * Phi(Y1 + A) = 0.05 >> B * Phi(Y2 - A) + (1-B) * Phi(Y2 + A) = 0.01 >> >> Y1 <

Re: [R] bug(?) in str() with strict.width = "cut" when appliedtodataframe with numeric component AND factor or character component withlongerlevels/strings

On 13-10-16 4:59 AM, Gerrit Eichner wrote: Dear Duncan, unfortunately, I have to correct myself in that I _can_ reproduce the problem after changing the global width-option to 70, say: Using the data frame X from before with the 'factory-fresh' setting for width and executing str( X, strict.w

Re: [R] bug(?) in str() with strict.width = "cut" when appliedtodataframe with numeric component AND factor or character component withlongerlevels/strings

On 13-10-16 4:59 AM, Gerrit Eichner wrote: Dear Duncan, unfortunately, I have to correct myself in that I _can_ reproduce the problem after changing the global width-option to 70, say: Using the data frame X from before with the 'factory-fresh' setting for width and executing str( X, strict.w

Re: [R] binding matrices

?rbind ?cbind Sent from my iPad On Oct 16, 2013, at 4:57, Mohammad Goodarzi wrote: > I have several matrices with the same size. I want to bind them one after > another. > Can you please let me know how to do it? > > Best Regards, > Mohammad > >[[alternative HTML version deleted]] > > __

[R] binding matrices

I have several matrices with the same size. I want to bind them one after another. Can you please let me know how to do it? Best Regards, Mohammad [[alternative HTML version deleted]] __ R-help@r-project.org mailing list https://stat.ethz.ch/ma

Re: [R] bug(?) in str() with strict.width = "cut" when appliedtodataframe with numeric component AND factor or character component withlongerlevels/strings

Dear Duncan, unfortunately, I have to correct myself in that I _can_ reproduce the problem after changing the global width-option to 70, say: Using the data frame X from before with the 'factory-fresh' setting for width and executing str( X, strict.width = "cut") 'data.frame': 11 obs. o

[R] course: Bayesian Data Analysis with R and WinBUGS

Dear list members, Best greetings and apologies for cross-posting. There are available places for the course: "Bayesian Data Analysis with R and WinBUGS", please find the description of the course below. If you have any question don't hesitate to contact me. Best regards, Pablo

Re: [R] bug(?) in str() with strict.width = "cut" when appliedtodataframe with numeric component AND factor or character component withlongerlevels/strings

Thanks, Duncan, for the good (indirect) hint: after a restart of R the problem is -- fortunately :-) -- not reproducible anymore for me either. The R session had been running for a longer time and I recall doing some (system-related) things outside of R that may have interfered with it; I jus