Re: [R] attach

2009-10-18 Thread Bernardo Rangel Tura
On Wed, 2009-10-14 at 07:21 +0200, Christophe Dutang wrote:
 Hi all,
 
 I have a question regarding the memory usage for the attach function.  
 Say I have a data.frame inputdat that I create with read.csv.
 
 I would like to know what happens on the memory side when I use
 attach(inputdata)
 
 Is there a second allocation of memory for inputdata?
 
 Then I'm using eval on a expression which depends on the columns of  
 inputdata. Is it better not to use attach function?
 
 Thanks in advance
 
 Christophe

Well, if you attach a data.frame twice times, it use your memory twice
times.

I don't use attach I prefer with(data.frame, command)
-- 
Bernardo Rangel Tura, M.D,MPH,Ph.D
National Institute of Cardiology
Brazil

__
R-help@r-project.org mailing list
https://stat.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide http://www.R-project.org/posting-guide.html
and provide commented, minimal, self-contained, reproducible code.


Re: [R] How do I access with the name of a (passed) function

2009-10-18 Thread Duncan Murdoch

On 17/10/2009 7:26 AM, Ajay Shah wrote:

How would I do something like this:

f - function(x, g) {
  s - as.character(g)   # THIS DOES NOT WORK
  sprintf(The %s of x is %.0f\n, s, g(x))
}


Gabor showed you how to do it if you pass an expression which evaluates 
to a function.  If you want to pass an expression that returns a 
character string as below, use


if (is.character(g)) {
  s - g
  g - get(s, parent.frame()) # gets it from the caller's frame
}



f(c(2,3,4), median)
f(c(2,3,4), mean)

and get the results 


   The median of x is 3
   The mean of x is 3



Duncan Murdoch

__
R-help@r-project.org mailing list
https://stat.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide http://www.R-project.org/posting-guide.html
and provide commented, minimal, self-contained, reproducible code.


Re: [R] function to convert lm model to LaTeX equation

2009-10-18 Thread Frank E Harrell Jr

Ista Zahn wrote:

Dear list,
I've tried several times to wrap my head around the Design library,
without much success. It does some really nice things, but I'm often
uncomfortable because I don't understand exactly what it's doing.
Anyway, one thing I really like is the latex.ols() function, which
converts an R linear model formula to a LaTeX equation.

So, I started writing a latex.lm() function (not actually using
classes at this point, I just named it that for consistency). This
turned out to be easy enough for simple cases (see code below), but
now I'm wondering a) if anyone knows of existing functions that do
this (again, for lm() models, I know I'm reinventing the wheel in as
far as the Design library goes), or if not, b) if anyone has
suggestions for improving the function below.

Thanks,
Ista

### Function to create LaTeX formula from lm() model. Needs amsmath
package in LaTeX. ###

latex.lm - function(object, file=, math.env=c($,$),
estimates=none, abbreviate = TRUE, abbrev.length=8, digits=3) {
# Get and format IV names
  co - c(Int, names(object$coefficients)[-1])
  co.n -  gsub(p.*), , co)
if(abbreviate == TRUE) {
  co.n - abbreviate(gsub(p.*), , co), minlength=abbrev.length)
}
# Get and format DV
  m.y - strsplit((as.character(object$call[2])),  ~ )[[1]][1]
# Write coefficent labels
  b.x - paste(\\beta_{, co.n ,}, sep=)
# Write error term
  e - \\epsilon_i
# Format coefficint x variable terms
  m.x - sub(}Int,}, paste(b.x, co.n,  + , sep=, collapse=))
# If inline estimates convert coefficient labels to values
if(estimates == inline) {
m.x - sub(Int, ,
paste(round(object$coefficients,digits=digits), co.n,  + , sep=,
collapse=))
m.x - gsub(\\+ \\-, -, m.x)
  }
# Format regression equation
  eqn - gsub(:,  \\times , paste(math.env[1], m.y,  = ,
m.x, e, sep=))
# Write the opening math mode tag and the model
  cat(eqn, file=file)
# If separae estimates format estimates and write them below the model
  if(estimates == separate) {
est - gsub(:,  \\times , paste(b.x,  = ,
round(object$coefficients, digits=digits), , , sep=, collapse=))
cat(,  \n \\text{where }, substr(est, 1, (nchar(est)-2)), file=file)
  }
# Write the closing math mode tag
  cat(math.env[2], \n, file=file)
}

# END latex.lm

Xvar1 - rnorm(20)
Xvar2 - rnorm(20)
Xvar3 - factor(rep(c(A,B),10))
Y.var - rnorm(20)
D - data.frame(Xvar1, Xvar2, Xvar3, Y.var)

x1 - lm(Y.var ~ pol(Xvar1, 3) + Xvar2*Xvar3, data=D)
latex.lm(x1)



It's not reinventing the wheel, in the sense that you are not attempting 
to handle the most needed features (simplifying regression splines and 
factoring out interaction terms with brackets).  I don't think you 
followed the posting guide though.  You didn't state your exact problem 
with Design and you didn't include any code.  Also note that the Design 
package is replaced with the rms package although latex features have 
not changed.


--
Frank E Harrell Jr   Professor and Chair   School of Medicine
 Department of Biostatistics   Vanderbilt University

__
R-help@r-project.org mailing list
https://stat.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide http://www.R-project.org/posting-guide.html
and provide commented, minimal, self-contained, reproducible code.


Re: [R] function to convert lm model to LaTeX equation

2009-10-18 Thread Ista Zahn
On Sun, Oct 18, 2009 at 9:09 AM, Frank E Harrell Jr
f.harr...@vanderbilt.edu wrote:
 Ista Zahn wrote:

 Dear list,
 I've tried several times to wrap my head around the Design library,
 without much success. It does some really nice things, but I'm often
 uncomfortable because I don't understand exactly what it's doing.
 Anyway, one thing I really like is the latex.ols() function, which
 converts an R linear model formula to a LaTeX equation.

 So, I started writing a latex.lm() function (not actually using
 classes at this point, I just named it that for consistency). This
 turned out to be easy enough for simple cases (see code below), but
 now I'm wondering a) if anyone knows of existing functions that do
 this (again, for lm() models, I know I'm reinventing the wheel in as
 far as the Design library goes), or if not, b) if anyone has
 suggestions for improving the function below.

 Thanks,
 Ista

 ### Function to create LaTeX formula from lm() model. Needs amsmath
 package in LaTeX. ###

 latex.lm - function(object, file=, math.env=c($,$),
 estimates=none, abbreviate = TRUE, abbrev.length=8, digits=3) {
 # Get and format IV names
  co - c(Int, names(object$coefficients)[-1])
  co.n -  gsub(p.*), , co)
    if(abbreviate == TRUE) {
      co.n - abbreviate(gsub(p.*), , co), minlength=abbrev.length)
    }
 # Get and format DV
  m.y - strsplit((as.character(object$call[2])),  ~ )[[1]][1]
 # Write coefficent labels
  b.x - paste(\\beta_{, co.n ,}, sep=)
 # Write error term
  e - \\epsilon_i
 # Format coefficint x variable terms
  m.x - sub(}Int,}, paste(b.x, co.n,  + , sep=, collapse=))
 # If inline estimates convert coefficient labels to values
    if(estimates == inline) {
    m.x - sub(Int, ,
 paste(round(object$coefficients,digits=digits), co.n,  + , sep=,
 collapse=))
    m.x - gsub(\\+ \\-, -, m.x)
  }
 # Format regression equation
  eqn - gsub(:,  \\times , paste(math.env[1], m.y,  = ,
 m.x, e, sep=))
 # Write the opening math mode tag and the model
  cat(eqn, file=file)
 # If separae estimates format estimates and write them below the model
  if(estimates == separate) {
    est - gsub(:,  \\times , paste(b.x,  = ,
 round(object$coefficients, digits=digits), , , sep=, collapse=))
    cat(,  \n \\text{where }, substr(est, 1, (nchar(est)-2)),
 file=file)
  }
 # Write the closing math mode tag
  cat(math.env[2], \n, file=file)
 }

 # END latex.lm

 Xvar1 - rnorm(20)
 Xvar2 - rnorm(20)
 Xvar3 - factor(rep(c(A,B),10))
 Y.var - rnorm(20)
 D - data.frame(Xvar1, Xvar2, Xvar3, Y.var)

 x1 - lm(Y.var ~ pol(Xvar1, 3) + Xvar2*Xvar3, data=D)
 latex.lm(x1)


 It's not reinventing the wheel, in the sense that you are not attempting to
 handle the most needed features (simplifying regression splines and
 factoring out interaction terms with brackets).  I don't think you followed
 the posting guide though.  You didn't state your exact problem with Design
 and you didn't include any code.  Also note that the Design package is
 replaced with the rms package although latex features have not changed.

Thank you for your response Prof. Harrell. Sorry my original post
didn't meet the guidelines -- it was poorly worded I'm afraid. The
question was not about the Design package, but about how to represent
a lm() model as a LaTeX equation, and specifically whether anyone had
already written code for this task, and if not how the function I
wrote could be improved. Thank you for you're suggestions about
needing to handle regression splines and factoring out interaction
terms, that's very helpful.

Thanks again,
Ista

 --
 Frank E Harrell Jr   Professor and Chair           School of Medicine
                     Department of Biostatistics   Vanderbilt University




-- 
Ista Zahn
Graduate student
University of Rochester
Department of Clinical and Social Psychology
http://yourpsyche.org

__
R-help@r-project.org mailing list
https://stat.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide http://www.R-project.org/posting-guide.html
and provide commented, minimal, self-contained, reproducible code.


Re: [R] QQ plot

2009-10-18 Thread Peter Dalgaard

Madan Sigdel wrote:

Dear users
 
I have applied following for my works:
 

library(car)
x-scan()

1: 0.92545
2: 0.89321
3: 0.9846
4: 2.9
5: 0.85968
6: 5.2
7: 4.66
8: 1.18788
9: 1.07683
10: 1.07683
11: 8.38
12: 7.423
13: 0.972
14: 3.73
15: 1.06474
16: 1.48
17: 0.92876
18: 2.26493
19: 0.85696
20: 1.89313
21: 2.71
22: 5
23: 3.02
24: 0.90369
25: 8.81
26: 1.69466
27: 1.07055
28: 1.17077
29: 2.31647
30: 0.83481
31: 5.42
32: 9.68
33: 1.27294
34: 5.49
35: 2.48
36: 1.55876
37: 1.41419
38: 0.94503
39: 5.24
40: 
Read 39 items

x-rgamma(39,shape=1.7,scale=1.6)
qq.plot(x,dist=gamma,shape=1.7,scale=1.6)



In this way can I get correct figure? My doubt is in the resulting

figure, gamma quantile are in X-axis and emperical are in Y-axis,
however the values in the Y-axis are up to 15. But my maximum point in
data set is 9.68. How can it be correct?

As far as I can see, it is not your data that you are plotting, but 
rgamma(..).



--
   O__   Peter Dalgaard Øster Farimagsgade 5, Entr.B
  c/ /'_ --- Dept. of Biostatistics PO Box 2099, 1014 Cph. K
 (*) \(*) -- University of Copenhagen   Denmark  Ph:  (+45) 35327918
~~ - (p.dalga...@biostat.ku.dk)  FAX: (+45) 35327907

__
R-help@r-project.org mailing list
https://stat.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide http://www.R-project.org/posting-guide.html
and provide commented, minimal, self-contained, reproducible code.


[R] optimization problem with constraints..

2009-10-18 Thread Prof. John C Nash
Apologies if this shows up a second time with uninformative header 
(apparently it got filtered, but ...), as I forgot to replace the 
subject line.


As a first try, use a bounds constrained method (L-BFGS-B or one from 
the r-forge Optimizer project 
http://r-forge.r-project.org/R/?group_id=395) and then add a penalty or 
barrier function to your objective function to take care of the

x1+x2  1  (the other end is implicit in the lower bounds on x1 and x2).

e.g.,   - const * log(1-x1-x2)

You should provide a feasible starting point. const scales the penalty.

Cheers, JN



 Message: 27
 Date: Sat, 17 Oct 2009 13:50:10 -0700 (PDT)
 From: kathie kathryn.lord2...@gmail.com
 Subject: [R]  optimization problem with constraints...
 To: r-help@r-project.org
 Message-ID: 25941686.p...@talk.nabble.com
 Content-Type: text/plain; charset=us-ascii


 Dear R users,
 I need some advises on how to use R to optimize a nonlinear function with
 the following constraints.

 f(x1,x2,x3,x4,x5,x6)
 s.t
 0  x1  1
 0  x2  1
 0  x1+x2  1
 -inf  x3  inf
 -inf  x4  inf
 0  x5  inf
 0  x6  inf

 Is there any built-in function or something for these constraint??

 Any suggestion will be greatly appreciated.

 Regards,

 Kathryn Lord -- View this message in context: 
http://www.nabble.com/optimization-problem-with-constraints...-tp25941686p25941686.html 
Sent from the R help mailing list archive at Nabble.com.


__
R-help@r-project.org mailing list
https://stat.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide http://www.R-project.org/posting-guide.html
and provide commented, minimal, self-contained, reproducible code.


Re: [R] Putting names on a ggplot

2009-10-18 Thread John Kane
Thanks Stefan, the annotate approach works beautifully.  I had not got that far 
in Hadley's book apparently :(

I'm not convinced though that the explaination

 you shouldn't use aes in this case since nampost,
 temprange, ... are not
 part of the dataframe year.

makes sense since it seems to work in this case unless I am missing something 
obvious. Mind you I'm good at missing the obvious especially in ggplot.

Example
=
library(ggplot2)

month  -  rep(month.abb[1:12], each=30)
   days  - rep(1:30, 12)
   temps  -  rnorm(length(month), mean=25,sd=8)
duration  - 1:length(month)

timedata - data.frame(month, days, temps, duration)
head(timedata)

mbs  - c(1,seq(30, 360,by=30))
namposts - mbs[1:12]
mlabs - month.name[1:12]
trange - range(timedata$temps)
drange - range(duration)

 
p  - ggplot(timedata, aes(duration, temps, colour=month)) + geom_line() +
   opts(legend.position = none, title=Yearly temperatures,
   axis.text.x = theme_blank(), axis.ticks = theme_blank())  


p  - p + geom_vline(xintercept= mbs) + 
  ylab(Temperature (C)) + xlab(Daily Temperatures) +
  geom_text(aes(x = namposts+2.5, y = trange[2], label = mlabs),
  data = timedata, size = 2.5, colour='black', hjust = 0, vjust = 0)
p


=
--- On Sat, 10/17/09, m...@z107.de m...@z107.de wrote:

 From: m...@z107.de m...@z107.de
 Subject: Re: [R] Putting names on a ggplot
 To: John Kane jrkrid...@yahoo.ca
 Cc: R R-help r-h...@stat.math.ethz.ch
 Received: Saturday, October 17, 2009, 5:53 PM
 hi,
 
 On Sat, Oct 17, 2009 at 02:04:43PM -0700, John Kane wrote:
  Putting names on a ggplot
  
  p - p + geom_text(aes(x = namposts + 2.5, y =
 temprange[2], label =
  mlabs),
        data = year, size =
 2.5, colour='black', hjust = 0, vjust = 0)
  
 
 you shouldn't use aes in this case since nampost,
 temprange, ... are not
 part of the dataframe year.
 
 It should also work with geom_text i guess, but I prefere
 annotate for
 thinks like that:
 
 p + annotate(text,x=namposts+2.5, y = temprange[2],
 label=
     mlabs,size=2.5,colour='black', hjust = 0,
 vjust = 0)
 
 regards,
     stefan
 


  __
Yahoo! Canada Toolbar: Search from anywhere on the web, and bookmark your 
favourite sites. Download it now
http://ca.toolbar.yahoo.com.

__
R-help@r-project.org mailing list
https://stat.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide http://www.R-project.org/posting-guide.html
and provide commented, minimal, self-contained, reproducible code.


Re: [R] looking for reference that covers convergence in distribution

2009-10-18 Thread Ben Bolker



Peng Yu wrote:
 
 On Sat, Oct 17, 2009 at 3:28 PM, Peter Dalgaard
 p.dalga...@biostat.ku.dk wrote:
 Ben Bolker wrote:


 Peng Yu wrote:

 I am looking for a good probability book that describes convergence in
 distribution. I have looked through Introduction to Probability by
 Charles M. Grinstead, J. Laurie Snell, but I don't find any formal
 description on convergence in distribution. Could somebody recommend a
 good book that cover this topic? Thank you!



 This mailing list is for R help, not general statistics help. May I
 respectfully request that you
 take your questions to a statistics help list instead?


 You may want to check out the R package ConvergenceConcepts, though.
 Supporting article due to appear in the next issue of the R Journal.
 
 I have checked sci.stat.math before my original post. But it is
 seriously flooded with junk posts. Before this problem is fixed, it is
 probably not very helpful to post anything there.
 
 I know my question is rudimentary. There are so many probability and
 statistics textbook online, and it is difficult for me to figure out
 which one fits my need. If you happen to know which book is the best
 for me to learn convergence in distribution, please let me know. Thank
 you!
 
 

I wonder if All of Statistics by Wasserman would be useful. It's uneven
but seems to be a pretty good whirlwind review of various topics.
-- 
View this message in context: 
http://www.nabble.com/looking-for-reference-that-covers-convergence-in-distribution-tp25940017p25947801.html
Sent from the R help mailing list archive at Nabble.com.

__
R-help@r-project.org mailing list
https://stat.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide http://www.R-project.org/posting-guide.html
and provide commented, minimal, self-contained, reproducible code.


Re: [R] R-help Digest, Vol 80, Issue 18

2009-10-18 Thread Prof. John C Nash
As a first try, use a bounds constrained method (L-BFGS-B or one from 
the r-forge Optimizer project 
http://r-forge.r-project.org/R/?group_id=395) and then add a penalty or 
barrier function to your objective function to take care of the

x1+x2  1  (the other end is implicit in the lower bounds on x1 and x2).

e.g.,   - const * log(1-x1-x2)

You should provide a feasible starting point. const scales the penalty.

Cheers, JN




Message: 27
Date: Sat, 17 Oct 2009 13:50:10 -0700 (PDT)
From: kathie kathryn.lord2...@gmail.com
Subject: [R]  optimization problem with constraints...
To: r-help@r-project.org
Message-ID: 25941686.p...@talk.nabble.com
Content-Type: text/plain; charset=us-ascii


Dear R users, 


I need some advises on how to use R to optimize a nonlinear function with
the following constraints.

f(x1,x2,x3,x4,x5,x6) 

s.t 


0  x1  1
0  x2  1
0  x1+x2  1
-inf  x3  inf
-inf  x4  inf
0  x5  inf
0  x6  inf

Is there any built-in function or something for these constraint??

Any suggestion will be greatly appreciated.

Regards,

Kathryn Lord 
-- View this message in context: http://www.nabble.com/optimization-problem-with-constraints...-tp25941686p25941686.html Sent from the R help mailing list archive at Nabble.com.


__
R-help@r-project.org mailing list
https://stat.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide http://www.R-project.org/posting-guide.html
and provide commented, minimal, self-contained, reproducible code.


Re: [R] Proper syntax for using varConstPower in nlme

2009-10-18 Thread Kingsford Jones
On Fri, Oct 16, 2009 at 3:56 PM, Michael A. Gilchrist mi...@utk.edu wrote:
 Hi Dieter,

 Thanks for the reply.  I had played with the initial conditions, but
 apparently not enough.  I finally found some that avoided the singularity
 issue.  Many thanks.

 More generally, I went over the documentation yet again in PB and I still
 find it a bit confusing.  They talk about using form = ~fitted(.) when
 discussing varPower, but the rest of the documentation seems to suggest that
 form = ~... should be used to indicate which covariate you assume the
 variance changes with.

 Could you or someone else provide some clarification?

I don't have PB in front of me, but see the 'form' agument definition
on the help page for any of the variance structures depending on a
covariate (varPower, varConstPower or varExp).  The ~fitted(.) and
~resid(.) notation specify that you would like the variance covariate
to be a function of the model being fit (the fitted values or
residuals), in which case the variance parameter \delta and the model
body are estimated iteratively.  Conversely, if you specify a constant
variance covariate such as ~age, there is no need for updating of the
variance covariate during optimization.

hth,

Kingsford Jones


 Thanks.

 Mike

 On Fri, 16 Oct 2009, Dieter Menne wrote:




 Michael A. Gilchrist wrote:

 -

  nlme(Count ~ quad.PBMC.model(aL, aN, T0),

 +   data = tissueData,
 +   weights = varConstPower(form =~ Count),
 +   start = list( fixed = c(rep(1000, 8), -2, -2) ),
 +   fixed = list(T0 ~ TypeTissue-1, aL ~ 1, aN ~ 1),
 +   random = aL + aN ~ 1|Tissue
 +   )
 Error in MEestimate(nlmeSt, grpShrunk) :
   Singularity in backsolve at level 0, block 1




 You could use varPower(form=~fitted()) (the default, also varPower()). In
 my
 experience this runs into some limitation quickly with nlme, because some
 boundary conditions make convergence fail.

 Try varPower(fixed = 0.5) first and play with the number.

 You should only use varConstPower when you have problems with values that
 cover a large range, coming close to zero, which could make varPower go
 havoc.

 Always do a plot of the result; the default plot gives you residual, and
 some indication how to proceed.

 Dieter


 --
 View this message in context:
 http://www.nabble.com/Proper-syntax-for-using-varConstPower-in-nlme-tp25914277p25927578.html
 Sent from the R help mailing list archive at Nabble.com.

 __
 R-help@r-project.org mailing list
 https://stat.ethz.ch/mailman/listinfo/r-help
 PLEASE do read the posting guide
 http://www.R-project.org/posting-guide.html
 and provide commented, minimal, self-contained, reproducible code.


 __
 R-help@r-project.org mailing list
 https://stat.ethz.ch/mailman/listinfo/r-help
 PLEASE do read the posting guide http://www.R-project.org/posting-guide.html
 and provide commented, minimal, self-contained, reproducible code.


__
R-help@r-project.org mailing list
https://stat.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide http://www.R-project.org/posting-guide.html
and provide commented, minimal, self-contained, reproducible code.


[R] Fit copulas to data

2009-10-18 Thread emkayenne

Hello!
For those of you who have ever dealt with copulas in R, youcould maybe help
me. :

I have used R to fit a couple of bivariate Archimedean copulas to financial
data. R gives a parameter and a z-value and a third number that is
supposedly some kind of p-value. 
An example of what I get after fitting a gumbel copula:

  Estimate Std. Error  z value Pr(|z|)
param 1.636907 0.07953911 20.579900
The maximized loglikelihood is  333.3923 
The convergence code is  0 

Who can tell me what the std error and the value of Pr(|z|) mean?

I would guess that the fit is pretty bad due to the small p-value, but I
think what is tested here is the Null that the parameter is 0. Clearly, this
Null would be rejected as the p value is small. But I am not sure what this
outcome really 

Couls somebody explain...? just give an interpretation of the outcome if the
fitCopula command?

Thank you so much,

Emkay
-- 
View this message in context: 
http://www.nabble.com/Fit-copulas-to-data-tp25948717p25948717.html
Sent from the R help mailing list archive at Nabble.com.

__
R-help@r-project.org mailing list
https://stat.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide http://www.R-project.org/posting-guide.html
and provide commented, minimal, self-contained, reproducible code.


Re: [R] Invoking par(mfrow...) on an already subdivided plot

2009-10-18 Thread Uwe Ligges



Maxwell Reback wrote:

I'd like to generate on a single device multiple plots, each of which
contains two plots. Essentially, I've got sub-plots which consist of
two tracks, the upper one displaying gene expression data, and the
lower one mapping position. I'd like to display four of these
two-track sub-plots on one device, but I can't seem to invoke the
par(mfrow=...) or layout(matrix(...)) functions at more than one
level.

I've got something like:

plot.gene-function(gene){
par(fig=c(0,1,0.3,1))  #expression: track 1
plot(...)
par(fig=c(0,1,0,0.4),new=TRUE) #position: track 2
plot(...)}

plot.multigene-function(gene,...){
pdf(paste(gene,.pdf,sep=))
par(mfrow=c(2,2))
tapply(gene,gene, plot.gene)
dev.off()}

The par(mfrow=) in plot.multigene, even when 'new=TRUE', is
disregarded and I just get one sub-plot per page. Suggestions?



You are right. You can use grid or other packages in order to use 
recursive kinds of subdivisions.


Uwe Ligges



(Thanks!)

__
R-help@r-project.org mailing list
https://stat.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide http://www.R-project.org/posting-guide.html
and provide commented, minimal, self-contained, reproducible code.


__
R-help@r-project.org mailing list
https://stat.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide http://www.R-project.org/posting-guide.html
and provide commented, minimal, self-contained, reproducible code.


[R] How to plot multiple data sets with different colors (also with legend)?

2009-10-18 Thread Peng Yu
The following commands only show the data in 'y'. I'm wondering how to
show the data in 'x' as well. I also want to add a legend to show that
blue points corresponds to 'x' and yellow points correspond to 'y'.
Could somebody let me know what the correct commands should be?

x=rbind(c(10,11),c(10,11))
y=cbind(-1:0,-1:0)
plot(y,col='yellow')
points(x,col='blue')

__
R-help@r-project.org mailing list
https://stat.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide http://www.R-project.org/posting-guide.html
and provide commented, minimal, self-contained, reproducible code.


[R] How to create MULTILEVELS in a dataset??

2009-10-18 Thread saurav pathak
Dear R users

I have a data set which has five variables. One depenedent variable y, and 4
Independent variables (education-level, householdincome, countrygdp and
countrygdpsquare). The first two are data corresponding to the individual
and the next two coorespond to the country to which the individual belongs
to. My data set does not make this distinction between individual level and
country level. Is there a way such that I can make R make countrygdp and
countrygdpsquare at a different level than the individual level data. In
other words I wish to transform my dataset such that it recognizes two
individual level variables to be at Level-1 and the other two country level
variables at Level-2.

I need to run a multilevel model, but first I must make my dataset recognise
data at Level-1 and Level-2. How can I create this country level group (gdp
and gdp^2) such that I can perform a multilevel model as follows:

lmer(y ~ education-level + householdincome + countrygdp + countrygdpsquare +
(1 I Level2),family=binomial(link=probit),data=dataset)

Please kindly help me with the relevant commands for creating this Level2
(having two variables)

Thanks
Saurav





Dr.Saurav Pathak
PhD, Univ.of.Florida
Mechanical Engineering
Doctoral Student
Innovation and Entrepreneurship
Imperial College Business School
s.patha...@imperial.ac.uk
0044-7795321121

[[alternative HTML version deleted]]

__
R-help@r-project.org mailing list
https://stat.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide http://www.R-project.org/posting-guide.html
and provide commented, minimal, self-contained, reproducible code.


Re: [R] Putting names on a ggplot

2009-10-18 Thread ml
hello,

On Sun, Oct 18, 2009 at 08:29:19AM -0700, John Kane wrote:
 Thanks Stefan, the annotate approach works beautifully.  I had not got
 that far in Hadley's book apparently :(
 
 I'm not convinced though that the explaination
 
  you shouldn't use aes in this case since nampost,
  temprange, ... are not
  part of the dataframe year.
 
 makes sense since it seems to work in this case unless I am missing
 something obvious. Mind you I'm good at missing the obvious especially
 in ggplot.


hmm, I thought that aes() is used to map variables from the used dataset
(data=...) to parts of the plot. But obviously this is wrong, since your
example below works.


In my (maybe out of date) pdf version of Hadley's book an page 46 it
says:

|Any variable in an aes() specification must be contained inside the
|plot or
|layer data. This is one of the ways in which ggplot2 objects are
|guaranteed
|to be entirely self-contained, so that they can be stored and
|re-used.

so, maybe the problem with your version is, you map variables of
your environment (namposts, trange) to aesthetic, which works fine in
your current environment, but fails if you save the plot and load it in
another environmento.

| Error in eval(expr, envir, enclos) : object 'namposts' not found


If you don't use save() on plots (I do not) this maybe not a problem for
you.

regards,
 stefan



 
 Example
 =
 library(ggplot2)
 
 month  -  rep(month.abb[1:12], each=30)
days  - rep(1:30, 12)
temps  -  rnorm(length(month), mean=25,sd=8)
 duration  - 1:length(month)
 
 timedata - data.frame(month, days, temps, duration)
 head(timedata)
 
 mbs  - c(1,seq(30, 360,by=30))
 namposts - mbs[1:12]
 mlabs - month.name[1:12]
 trange - range(timedata$temps)
 drange - range(duration)
 
  
 p  - ggplot(timedata, aes(duration, temps, colour=month)) + geom_line() +
opts(legend.position = none, title=Yearly temperatures,
axis.text.x = theme_blank(), axis.ticks = theme_blank())  
 
 
 p  - p + geom_vline(xintercept= mbs) + 
   ylab(Temperature (C)) + xlab(Daily Temperatures) +
   geom_text(aes(x = namposts+2.5, y = trange[2], label = mlabs),
   data = timedata, size = 2.5, colour='black', hjust = 0, vjust = 0)
 p
 
 
 =
 --- On Sat, 10/17/09, m...@z107.de m...@z107.de wrote:
 
  From: m...@z107.de m...@z107.de
  Subject: Re: [R] Putting names on a ggplot
  To: John Kane jrkrid...@yahoo.ca
  Cc: R R-help r-h...@stat.math.ethz.ch
  Received: Saturday, October 17, 2009, 5:53 PM
  hi,
  
  On Sat, Oct 17, 2009 at 02:04:43PM -0700, John Kane wrote:
   Putting names on a ggplot
   
   p - p + geom_text(aes(x = namposts + 2.5, y =
  temprange[2], label =
   mlabs),
         data = year, size =
  2.5, colour='black', hjust = 0, vjust = 0)
   
  
  you shouldn't use aes in this case since nampost,
  temprange, ... are not
  part of the dataframe year.
  
  It should also work with geom_text i guess, but I prefere
  annotate for
  thinks like that:
  
  p + annotate(text,x=namposts+2.5, y = temprange[2],
  label=
      mlabs,size=2.5,colour='black', hjust = 0,
  vjust = 0)
  
  regards,
      stefan
  
 
 
   __
 Yahoo! Canada Toolbar: Search from anywhere on the web, and bookmark your 
 favourite sites. Download it now
 http://ca.toolbar.yahoo.com.


__
R-help@r-project.org mailing list
https://stat.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide http://www.R-project.org/posting-guide.html
and provide commented, minimal, self-contained, reproducible code.


Re: [R] using a custom color sequence for image()

2009-10-18 Thread Jim Lemon

On 10/17/2009 01:25 AM, Rajarshi Guha wrote:

Hi, I'd like to use a custom color sequence (black - low values, green -
high values) in am image() plot. While I can specify colors (say a  sequence
of grays) to the col argument, the ordering is getting messed up. I have two
questions:

1. How can I get a sequence of say 256 colors starting from black and ending
in green?
2. How is this specified to image() such that it uses the colors in the
proper ordering?
   

Hi Rajarshi,
Adding to what Albert wrote, if you mean that you want black to gray and 
then gray to green, you can get it by creating two sequences of colors 
and then joining them together:


ccolors-c(rgb(seq(0,0.5,length.out=128),
 seq(0,0.5,length.out=128),seq(0,0.5,length.out=128)),
 rgb(seq(0.5,0,length.out=128),seq(0.5,1,length.out=128),
 seq(0.5,0,length.out=128)))
image(...,col=ccolors,...)

See the examples in color2D.matplot for more information.

Jim

__
R-help@r-project.org mailing list
https://stat.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide http://www.R-project.org/posting-guide.html
and provide commented, minimal, self-contained, reproducible code.


Re: [R] How to plot multiple data sets with different colors (also with legend)?

2009-10-18 Thread Matthieu Dubois
Hi, 

the blue point is not shown simply because it is printed outside 
the current plot area. If you want to use the base graphics, you 
have to manually define the xlim and ylim of the plot. Legend is added 
with the command legend. 

E.g.
x=rbind(c(10,11),c(10,11))
y=cbind(-1:0,-1:0)
plot(y,col='yellow', xlim=c(-1,11), ylim=c(-1,11))
points(x,col='blue')
legend(topleft, c(x,y), col=c('blue', 'yellow'), pch=1)

This is nevertheless most easily done in ggplot2. 
E.g.
library(ggplot2)
# put the whole data in a data frame
# and add a new variable to distinguish both
dat - data.frame(rbind(x,y), var=rep(c('x','y'), each=2))
qplot(x=X1,y=X2, colour=var, data=dat)


HTH, 

Matthieu

__
R-help@r-project.org mailing list
https://stat.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide http://www.R-project.org/posting-guide.html
and provide commented, minimal, self-contained, reproducible code.


Re: [R] Putting names on a ggplot

2009-10-18 Thread John Kane


--- On Sun, 10/18/09, m...@z107.de m...@z107.de wrote:

 From: m...@z107.de m...@z107.de
 Subject: Re: [R] Putting names on a ggplot
 To: John Kane jrkrid...@yahoo.ca
 Cc: R R-help r-h...@stat.math.ethz.ch
 Received: Sunday, October 18, 2009, 6:05 PM
 hello,
 
 On Sun, Oct 18, 2009 at 08:29:19AM -0700, John Kane wrote:
  Thanks Stefan, the annotate approach works
 beautifully.  I had not got
  that far in Hadley's book apparently :(
  
  I'm not convinced though that the explaination
  
   you shouldn't use aes in this case since
 nampost,
   temprange, ... are not
   part of the dataframe year.
  
  makes sense since it seems to work in this case unless
 I am missing
  something obvious. Mind you I'm good at missing the
 obvious especially
  in ggplot.
 
 
 hmm, I thought that aes() is used to map variables from the
 used dataset
 (data=...) to parts of the plot. But obviously this is
 wrong, since your
 example below works.
 
 
 In my (maybe out of date) pdf version of Hadley's book an
 page 46 it
 says:
 
 |Any variable in an aes() specification must be contained
 inside the
 |plot or
 |layer data. This is one of the ways in which ggplot2
 objects are
 |guaranteed
 |to be entirely self-contained, so that they can be stored
 and
 |re-used.

I just realized I deleted Hadley's pdf so I cannot check this but it 
sounds right.

 
 so, maybe the problem with your version is, you map
 variables of
 your environment (namposts, trange) to aesthetic, which
 works fine in
 your current environment, but fails if you save the plot
 and load it in
 another environmento.

 
 | Error in eval(expr, envir, enclos) : object 'namposts'
 not found
 
 
 If you don't use save() on plots (I do not) this maybe not
 a problem for
 you.
 

That's interesting. We're getting different error messages. I get

 p
Error in data.frame(..., check.names = FALSE) : 
  arguments imply differing number of rows: 12, 366

which makes your first idea seem more reasonable but I don't see why my other 
example would work if it was a data.frame problem :(


Thanks for following up on this.  

I just installed 2.9.2 yesterday so I think I'm up to date on ggplot2 .



  
  Example
 
 =
  library(ggplot2)
  
  month  -  rep(month.abb[1:12], each=30)
         days  - rep(1:30,
 12)
         temps  - 
 rnorm(length(month), mean=25,sd=8)
  duration  - 1:length(month)
  
  timedata - data.frame(month, days, temps,
 duration)
  head(timedata)
  
  mbs  - c(1,seq(30, 360,by=30))
  namposts - mbs[1:12]
  mlabs - month.name[1:12]
  trange - range(timedata$temps)
  drange - range(duration)
  
               
  p  - ggplot(timedata, aes(duration, temps,
 colour=month)) + geom_line() +
         opts(legend.position =
 none, title=Yearly temperatures,
         axis.text.x =
 theme_blank(), axis.ticks = theme_blank())  
  
  
  p  - p + geom_vline(xintercept= mbs) + 
        ylab(Temperature (C))
 + xlab(Daily Temperatures) +
        geom_text(aes(x =
 namposts+2.5, y = trange[2], label = mlabs),
        data = timedata, size =
 2.5, colour='black', hjust = 0, vjust = 0)
  p
  
  
 
 =
  --- On Sat, 10/17/09, m...@z107.de m...@z107.de wrote:
  
   From: m...@z107.de m...@z107.de
   Subject: Re: [R] Putting names on a ggplot
   To: John Kane jrkrid...@yahoo.ca
   Cc: R R-help r-h...@stat.math.ethz.ch
   Received: Saturday, October 17, 2009, 5:53 PM
   hi,
   
   On Sat, Oct 17, 2009 at 02:04:43PM -0700, John
 Kane wrote:
Putting names on a ggplot

p - p + geom_text(aes(x = namposts +
 2.5, y =
   temprange[2], label =
mlabs),
          data = year, size =
   2.5, colour='black', hjust = 0, vjust = 0)

   
   you shouldn't use aes in this case since
 nampost,
   temprange, ... are not
   part of the dataframe year.
   
   It should also work with geom_text i guess, but I
 prefere
   annotate for
   thinks like that:
   
   p + annotate(text,x=namposts+2.5, y =
 temprange[2],
   label=
       mlabs,size=2.5,colour='black', hjust = 0,
   vjust = 0)
   
   regards,
       stefan
   
  
  
    
    __
  Yahoo! Canada Toolbar: Search from anywhere on the
 web, and bookmark your favourite sites. Download it now
  http://ca.toolbar.yahoo.com.
  
 


  __
Ask a question on any topic and get answers from real people. Go to Yahoo! Answe
__
R-help@r-project.org mailing list
https://stat.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide http://www.R-project.org/posting-guide.html
and provide commented, minimal, self-contained, reproducible code.


Re: [R] How to plot multiple data sets with different colors (also with legend)?

2009-10-18 Thread Jim Lemon

On 10/19/2009 07:37 AM, Peng Yu wrote:

The following commands only show the data in 'y'. I'm wondering how to
show the data in 'x' as well. I also want to add a legend to show that
blue points corresponds to 'x' and yellow points correspond to 'y'.
Could somebody let me know what the correct commands should be?

x=rbind(c(10,11),c(10,11))
y=cbind(-1:0,-1:0)
plot(y,col='yellow')
points(x,col='blue')

   

Hi Peng,
To show the x points, you will have to set both the xlim and ylim 
arguments:


plot(y,col=yellow,xlim=c(-1,11),ylim=c(-1,11))
points(x,col=blue)

I'm not sure why you are passing the points as matrices, but this means 
that the x points are the same.


Jim

__
R-help@r-project.org mailing list
https://stat.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide http://www.R-project.org/posting-guide.html
and provide commented, minimal, self-contained, reproducible code.


[R] rbind to array members

2009-10-18 Thread Another Oneforyou

Hi,
I would like to add rows to arbitrary tables within a 3dimensional array.
I can directly add data to an existing row of a table:
 x - array(0,c(1,3,2)) x[,,1] - c(1,2,3)
And I can even add a row to the table and assign to another object.
 y - rbind(x[,,1], c(4,5,6))
and 'y' is what I want it to be: y     [,1] [,2] [,3][1,]    1    2   3[2,]    
4    5   6
but I can't do this within the 3dimensional array:
 x[,,1] - rbind(x[,,1], c(4,5,6))Error in x[, , 1] - rbind(x[, , 1], c(4, 5, 
 6)) :   number of items to replace is not a multiple of replacement length
Does anyone know I can add rows to one of tables within the array?
Thanks.

  
_
Hotmail: Free, trusted and rich email service.

__
R-help@r-project.org mailing list
https://stat.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide http://www.R-project.org/posting-guide.html
and provide commented, minimal, self-contained, reproducible code.


[R] CRAN (and crantastic) updates this week

2009-10-18 Thread Crantastic
CRAN (and crantastic) updates this week

New packages


* adaptTest (1.0)
  Marc Vandemeulebroecke
  http://crantastic.org/packages/adaptTest

  The functions defined in this program serve for implementing adaptive
  two-stage tests. Currently, four tests are included: Bauer and
  Koehne (1994), Lehmacher and Wassmer (1999), Vandemeulebroecke
  (2006), and the horizontal conditional error function. User-defined
  tests can also be implemented. Reference: Vandemeulebroecke, An
  investigation of two-stage tests, Statistica Sinica 2006.

* bcv (1.0)
  Patrick O. Perry
  http://crantastic.org/packages/bcv

  This package implements methods for choosing the rank of an SVD
  approximation via cross validation.  It provides both Gabriel-style
  block holdouts and Wold-style speckled holdouts.  Also included
  is an implementation of the SVDImpute algorithm.  For more
  information about Bi-cross-validation, see Owen  Perry's 2009 AOAS
  article (at http://arxiv.org/abs/0908.2062) and Perry's 2009 PhD
  thesis (at http://arxiv.org/abs/0909.3052).

* dse1 (2009.10-1)
  Paul Gilbert
  http://crantastic.org/packages/dse1

  This package is only to aid transition to the unbundled dse package.
  It has no functions, but simply requires package EvalEst.

* dse2 (2009.10-1)
  Paul Gilbert
  http://crantastic.org/packages/dse2

  This package is only to aid transition to the unbundled dse package.
  It has no functions, but simply requires package EvalEst.

* EvalEst (2009.10-2)
  Paul Gilbert
  http://crantastic.org/packages/EvalEst

  Multivariate Time Series - extensions.See ?00dse-Intro for more
  details.

* glmdm (0.51)
  Jeff Gill
  http://crantastic.org/packages/glmdm

  R CODE FOR SIMULATION OF GLMDM

* integrativeME (1.1)
  Kim-Anh Le Cao
  http://crantastic.org/packages/integrativeME

  Mixture of experts models (Jacobs et al., 1991) were introduced to
  account for nonlinearities and other complexities in the data. It is
  based on a divide-and-conquer strategy. Mixture of experts are of
  interest due to their wide applicability and the advantages of fast
  learning via the expectation-maximization (EM) algorithm. We have
  extended and implemented mixture of experts to combine categorical
  clinical factors and continuous microarray data in a binary
  classification framework to analyze cancer studies. To provide a
  hybrid signature of clinical factors and gene markers, we propose to
  apply different gene selection procedures as a first step.

* integrativeMEdata (1.0)
  Kim-Anh Le Cao
  http://crantastic.org/packages/integrativeMEdata

  This package contains data sets with matched categorical clinical
  factors and microarray data for three cancer studies. This is part
  of the integrativeME package that combines these two types of
  variables in a binary classification framework by selecting a hybrid
  signature of clinical factors and gene markers.

* latticedl (1.0)
  Toby Dylan Hocking
  http://crantastic.org/packages/latticedl

  Direct labeling functions that use the lattice package.

* nodeHarvest (0.1)
  Nicolai Meinshausen
  http://crantastic.org/packages/nodeHarvest

  Node harvest is a simple interpretable tree-like estimator for
  high-dimensional regression and classification. A few nodes are
  selected from an initially large ensemble of nodes, each associated
  with a positive weight. New observations can fall into one or
  several nodes and predictions are the weighted average response
  across all these groups. The package offers visualization of the
  estimator. Predictions can return the nodes a new observation fell
  into, along with the mean response of training observations in each
  node, offering a simple explanation of the prediction.

* sublogo (1.0)
  Toby Dylan Hocking
  http://crantastic.org/packages/sublogo

  Visualize correlation in biological sequence data using sublogo
  dendrogram plots.


Updated packages


adimpro (0.7.3), aplpack (1.2.2), approximator (1.1-6), aws (1.6-1),
aylmer (1.0-4), BACCO (2.0-4), BB (2009.9-1), binMto (0.0-4), boot
(1.2-41), bootspecdens (3.0), bqtl (1.0-25), CalciOMatic (1.1-3),
calibrator (1.1-7), ccgarch (0.1.7), choplump (1.0), CircStats
(0.2-4), clv (0.3-2), compare (0.2-3), condGEE (0.1-3), contfrac
(1.1-8), Davies (1.1-5), deSolve (1.5), Devore7 (0.7.2), dlnm (1.0.2),
dplR (1.1.9.4), drfit (0.05-95), dse (2009.10-1), dti (0.8-2), EDR
(0.6-3), effects (2.0-9), effects (2.0-10), eha (1.2-12), eiPack
(0.1-6), emulator (1.1-7), epicalc (2.9.2.8), FGN (1.2), FitAR (1.79),
forward (1.0.3), fts (0.7.6), gamlss (3.0-1), gamlss.cens (3.0.1),
gamlss.data (3.0-1), gamlss.dist (3.0-1), gamlss.mx (3.0-1), gamlss.nl
(3.0-1), gamlss.tr (3.0-1), geoR (1.6-27), geoRglm (0.8-26), gld
(1.8.4), glmmML (0.81-6), gplots (2.7.2), grouped (0.6-0), HAPim
(1.3), hash (1.0.2), hdrcde (2.12), heplots (0.8-10), intervals
(0.13.1), irtoys (0.1.2), isotone (0.8-7), ivivc (0.1.5), Kendall
(2.1), kernlab (0.9-9), ks (1.6.8), lcda (0.2), 

Re: [R] How to plot multiple data sets with different colors (also with legend)?

2009-10-18 Thread Peng Yu
On Sun, Oct 18, 2009 at 5:42 PM, Matthieu Dubois matth...@gmail.com wrote:
 Hi,

 the blue point is not shown simply because it is printed outside
 the current plot area. If you want to use the base graphics, you
 have to manually define the xlim and ylim of the plot. Legend is added
 with the command legend.

 E.g.
 x=rbind(c(10,11),c(10,11))
 y=cbind(-1:0,-1:0)
 plot(y,col='yellow', xlim=c(-1,11), ylim=c(-1,11))
 points(x,col='blue')
 legend(topleft, c(x,y), col=c('blue', 'yellow'), pch=1)

 This is nevertheless most easily done in ggplot2.
 E.g.
 library(ggplot2)
 # put the whole data in a data frame
 # and add a new variable to distinguish both
 dat - data.frame(rbind(x,y), var=rep(c('x','y'), each=2))
 qplot(x=X1,y=X2, colour=var, data=dat)

qplot generates a figure with some background grid. If I just want a
blank background (as in plot), what options should I specify? How to
specific the color like 'red' and 'blue' explicitly?

I have read the review for ggplot2 book on amazon. The rates are
unanimously high. I want to know how much effort I should spend to
learn ggplot2 versus conventional graphics R packages. Can ggplot2 do
all the graphics tasks? Is it much easier to learn than conventional
graphics packages?

__
R-help@r-project.org mailing list
https://stat.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide http://www.R-project.org/posting-guide.html
and provide commented, minimal, self-contained, reproducible code.


[R] Error: object 'cloud' not found

2009-10-18 Thread PerfectTiling

Hi,

I installed the lattice package so I can plot 3D cloud scatterplots:

   install.packages(lattice)

But (after successfully installing from the Berkeley mirror), R insists it
cannot find the cloud function, part of the lattice package:

   cloud
  Error: object 'cloud' not found

What did I do wrong?

Thanks!

PT
-- 
View this message in context: 
http://www.nabble.com/%22Error%3A-object-%27cloud%27-not-found%22-tp25952117p25952117.html
Sent from the R help mailing list archive at Nabble.com.

__
R-help@r-project.org mailing list
https://stat.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide http://www.R-project.org/posting-guide.html
and provide commented, minimal, self-contained, reproducible code.


Re: [R] Error: object 'cloud' not found

2009-10-18 Thread Rolf Turner


On 19/10/2009, at 2:16 PM, PerfectTiling wrote:



Hi,

I installed the lattice package so I can plot 3D cloud scatterplots:


install.packages(lattice)


But (after successfully installing from the Berkeley mirror), R  
insists it

cannot find the cloud function, part of the lattice package:


cloud

  Error: object 'cloud' not found

What did I do wrong?


You didn't ***load*** the package, i.e. you didn't do

library(lattice)

cheers,

Rolf Turner

##
Attention:\ This e-mail message is privileged and confid...{{dropped:9}}

__
R-help@r-project.org mailing list
https://stat.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide http://www.R-project.org/posting-guide.html
and provide commented, minimal, self-contained, reproducible code.


Re: [R] Error: object 'cloud' not found

2009-10-18 Thread Roy Mendelssohn

library(lattice)

Roy M.

On Oct 18, 2009, at 6:16 PM, PerfectTiling wrote:



Hi,

I installed the lattice package so I can plot 3D cloud scatterplots:


install.packages(lattice)


But (after successfully installing from the Berkeley mirror), R  
insists it

cannot find the cloud function, part of the lattice package:


cloud

Error: object 'cloud' not found

What did I do wrong?

Thanks!

PT
--
View this message in context: 
http://www.nabble.com/%22Error%3A-object-%27cloud%27-not-found%22-tp25952117p25952117.html
Sent from the R help mailing list archive at Nabble.com.

__
R-help@r-project.org mailing list
https://stat.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide http://www.R-project.org/posting-guide.html
and provide commented, minimal, self-contained, reproducible code.


**
The contents of this message do not reflect any position of the U.S.  
Government or NOAA.

**
Roy Mendelssohn
Supervisory Operations Research Analyst
NOAA/NMFS
Environmental Research Division
Southwest Fisheries Science Center
1352 Lighthouse Avenue
Pacific Grove, CA 93950-2097

e-mail: roy.mendelss...@noaa.gov (Note new e-mail address)
voice: (831)-648-9029
fax: (831)-648-8440
www: http://www.pfeg.noaa.gov/

Old age and treachery will overcome youth and skill.
From those who have been given much, much will be expected

__
R-help@r-project.org mailing list
https://stat.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide http://www.R-project.org/posting-guide.html
and provide commented, minimal, self-contained, reproducible code.


Re: [R] Error: object 'cloud' not found

2009-10-18 Thread Jorge Ivan Velez
Hi PT,
Try loading the lattice package first:

# install.packages('lattice')
require(lattice)
?cloud

HTH,
Jorge

On Sun, Oct 18, 2009 at 9:16 PM, PerfectTiling  wrote:


 Hi,

 I installed the lattice package so I can plot 3D cloud scatterplots:

   install.packages(lattice)

 But (after successfully installing from the Berkeley mirror), R insists it
 cannot find the cloud function, part of the lattice package:

   cloud
  Error: object 'cloud' not found

 What did I do wrong?

 Thanks!

 PT
 --
 View this message in context:
 http://www.nabble.com/%22Error%3A-object-%27cloud%27-not-found%22-tp25952117p25952117.html
 Sent from the R help mailing list archive at Nabble.com.

 __
 R-help@r-project.org mailing list
 https://stat.ethz.ch/mailman/listinfo/r-help
 PLEASE do read the posting guide
 http://www.R-project.org/posting-guide.html
 and provide commented, minimal, self-contained, reproducible code.


[[alternative HTML version deleted]]

__
R-help@r-project.org mailing list
https://stat.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide http://www.R-project.org/posting-guide.html
and provide commented, minimal, self-contained, reproducible code.


Re: [R] rbind to array members

2009-10-18 Thread Richard M. Heiberger

library(abind)  ## array binding

__
R-help@r-project.org mailing list
https://stat.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide http://www.R-project.org/posting-guide.html
and provide commented, minimal, self-contained, reproducible code.


[R] Why points() is defined specially for a 1 by 2 matrix?

2009-10-18 Thread Peng Yu
x=cbind(1:4,3:6)
png('one_point.png')
plot(x[1:3,],xlim=c(-1,11),ylim=c(-1,11),pch=1)
points(x[4,],pch=2)# this is plotted as two points
#although I meant only one point
legend(topleft, c(x,y),pch=c(1,2))
dev.off()

The above code will produce 5 points instead of 4 points. If I want to
have 4 points, I have to use the following code. But the below code is
a little bit tedious. I'm wondering if there is a way that I still use
the above code (with a little change) to generate 4 points?


x=cbind(1:4,3:6)
png('one_point_split.png')
plot(x[1:3,1],x[1:3,2],xlim=c(-1,11),ylim=c(-1,11),pch=1)
points(x[4,1],x[4,2],pch=2)
legend(topleft, c(x,y),pch=c(1,2))
dev.off()

__
R-help@r-project.org mailing list
https://stat.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide http://www.R-project.org/posting-guide.html
and provide commented, minimal, self-contained, reproducible code.


[R] Identifying similar but not identical rows in a dataframe

2009-10-18 Thread Jeffrey Salinger
I would like to identify _almost_ duplicated rows in a data frame.  For 
example, I might declare as duplicates pairs of rows that are alike at about 
80% of their columns.  When working with tens of thousands of rows and upwards 
of 20 columns an iterative approach, testing all permutations, can be time 
consuming. 

 Duplicated() with incomparables sounds like the ticket.  But previous 
discussion in this forum indicates that specifying an
incomparable value when using duplicated() on a data frame is not yet
implemented. 

Any suggestions about how to implement this efficiently would be appreciated.  

All data are numerical, and each datum could, for example, be reduced to a byte 
representation in a string.  A fuzzy matching approach with agrep() might be 
possible.

Thanks.



  __
Be smarter than spam. See how smart SpamGuard is at giving junk email the b

[[alternative HTML version deleted]]

__
R-help@r-project.org mailing list
https://stat.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide http://www.R-project.org/posting-guide.html
and provide commented, minimal, self-contained, reproducible code.


[R] Filtering on a dataframe- newbie question

2009-10-18 Thread ANJAN PURKAYASTHA
Hi,
newbie question. I have a data-frame with 3 named columns: Name, Obs1, Obs2.
The Name column members are made of alphanumeric characters: T1, T2, T3 etc.
I would like to acess only that subset of the data-frame with Name == T44.
X - dataframe[dataframe$Name=='T44'] does not work.
Any ideas on how to do this?
I'm sure I'm missing a simple concept here.

Thanks,
Anjan


-- 
=
anjan purkayastha, phd
bioinformatics analyst
whitehead institute for biomedical research
nine cambridge center
cambridge, ma 02142

purkayas [at] wi [dot] mit [dot] edu
703.740.6939

[[alternative HTML version deleted]]

__
R-help@r-project.org mailing list
https://stat.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide http://www.R-project.org/posting-guide.html
and provide commented, minimal, self-contained, reproducible code.


Re: [R] Filtering on a dataframe- newbie question

2009-10-18 Thread milton ruser
Hi there,

Try ?subset
?str may also be useful..

bests

milton

On Sun, Oct 18, 2009 at 11:10 PM, ANJAN PURKAYASTHA 
anjan.purkayas...@gmail.com wrote:

 Hi,
 newbie question. I have a data-frame with 3 named columns: Name, Obs1,
 Obs2.
 The Name column members are made of alphanumeric characters: T1, T2, T3
 etc.
 I would like to acess only that subset of the data-frame with Name == T44.
 X - dataframe[dataframe$Name=='T44'] does not work.
 Any ideas on how to do this?
 I'm sure I'm missing a simple concept here.

 Thanks,
 Anjan


 --
 =
 anjan purkayastha, phd
 bioinformatics analyst
 whitehead institute for biomedical research
 nine cambridge center
 cambridge, ma 02142

 purkayas [at] wi [dot] mit [dot] edu
 703.740.6939

[[alternative HTML version deleted]]

 __
 R-help@r-project.org mailing list
 https://stat.ethz.ch/mailman/listinfo/r-help
 PLEASE do read the posting guide
 http://www.R-project.org/posting-guide.htmlhttp://www.r-project.org/posting-guide.html
 and provide commented, minimal, self-contained, reproducible code.


[[alternative HTML version deleted]]

__
R-help@r-project.org mailing list
https://stat.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide http://www.R-project.org/posting-guide.html
and provide commented, minimal, self-contained, reproducible code.


Re: [R] How to plot multiple data sets with different colors (also with legend)?

2009-10-18 Thread Ista Zahn
Hi Peng,
Comments below.

On Sun, Oct 18, 2009 at 9:22 PM, Peng Yu pengyu...@gmail.com wrote:
 On Sun, Oct 18, 2009 at 5:42 PM, Matthieu Dubois matth...@gmail.com wrote:
 Hi,

 the blue point is not shown simply because it is printed outside
 the current plot area. If you want to use the base graphics, you
 have to manually define the xlim and ylim of the plot. Legend is added
 with the command legend.

 E.g.
 x=rbind(c(10,11),c(10,11))
 y=cbind(-1:0,-1:0)
 plot(y,col='yellow', xlim=c(-1,11), ylim=c(-1,11))
 points(x,col='blue')
 legend(topleft, c(x,y), col=c('blue', 'yellow'), pch=1)

 This is nevertheless most easily done in ggplot2.
 E.g.
 library(ggplot2)
 # put the whole data in a data frame
 # and add a new variable to distinguish both
 dat - data.frame(rbind(x,y), var=rep(c('x','y'), each=2))
 qplot(x=X1,y=X2, colour=var, data=dat)

 qplot generates a figure with some background grid. If I just want a
 blank background (as in plot), what options should I specify? How to
 specific the color like 'red' and 'blue' explicitly?


You can get a more traditional look by issuing

theme_set(theme_bw())

before the call to qplot(). The colors are controlled by the a scale,
which you can override as follows:

qplot(x=X1,y=X2, colour=var, data=dat) + scale_colour_manual(values =
c(red,green))

 I have read the review for ggplot2 book on amazon. The rates are
 unanimously high. I want to know how much effort I should spend to
 learn ggplot2 versus conventional graphics R packages. Can ggplot2 do
 all the graphics tasks? Is it much easier to learn than conventional
 graphics packages?

ggplot2 can do most things that can be done in base graphics. It makes
many things that are difficult in base easy, like faceting and mapping
variables to a wide variety of scales. I myself use ggplot2 almost
exclusively. I don't know base graphics at all, and I'm able to
accomplish all my graphing needs with ggplot2. I would not say its
easier than base graphics, just different. Some things are easier with
base graphics, other things are easier with ggplot. I use it because I
like the consistent and rational user interface (and the default theme
is nice to look at).

The place to start learning ggplot2 (while your're waiting for the
book to be shipped perhaps) is http://had.co.nz/ggplot2/.

-Ista

 __
 R-help@r-project.org mailing list
 https://stat.ethz.ch/mailman/listinfo/r-help
 PLEASE do read the posting guide http://www.R-project.org/posting-guide.html
 and provide commented, minimal, self-contained, reproducible code.




-- 
Ista Zahn
Graduate student
University of Rochester
Department of Clinical and Social Psychology
http://yourpsyche.org

__
R-help@r-project.org mailing list
https://stat.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide http://www.R-project.org/posting-guide.html
and provide commented, minimal, self-contained, reproducible code.


Re: [R] Why points() is defined specially for a 1 by 2 matrix?

2009-10-18 Thread Richard M. Heiberger

 points(x[4,],pch=2)# this is plotted as two points

drops what it sees as an unnecessary dimension.

Use

 points(x[4,, drop=FALSE], pch=2)

See FAQ 7.5

tmp - matrix(1:2)
tmp
tmp[,1]
tmp[,1,drop=FALSE]

__
R-help@r-project.org mailing list
https://stat.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide http://www.R-project.org/posting-guide.html
and provide commented, minimal, self-contained, reproducible code.


[R] modifying model coefficients

2009-10-18 Thread tdm

I have build a model but want to then manipulate the coefficients in some
way.

I can extract the coefficients and do the changes I need, but how do I then
put these new coefficients back in the model so I can use the predict
function? 

my_model - lm(x ~ . , data=my_data)
my_scores - predict(my_model, my_data)

my_coeffs - coef(my_model)

## here we manipulate my_coeffs
## and then want to set the my_model 
## coefficients to the new values so we 
## predict using the new values

my_model.coefs - my_coeffs ?? how is this done?

?? so that this will work with the new coefficients
my_scores_new - predict(my_model, my_data)

Any code snippets would be appreciated very much.

-- 
View this message in context: 
http://www.nabble.com/modifying-model-coefficients-tp25952879p25952879.html
Sent from the R help mailing list archive at Nabble.com.

__
R-help@r-project.org mailing list
https://stat.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide http://www.R-project.org/posting-guide.html
and provide commented, minimal, self-contained, reproducible code.


Re: [R] Filtering on a dataframe- newbie question

2009-10-18 Thread Rolf Turner


On 19/10/2009, at 4:23 PM, milton ruser wrote:


Hi there,

Try ?subset


No. Don't.

Just do:

X - dataframe[dataframe$Name=='T44',]

Note the comma in the penultimate position.

Read up on array indexing; see ?[ and An Introduction to R,
section 5.2 for starters.

A data frame is two dimensional; you were basically trying to
treat it as one dimensional.

And don't call your data frame ``dataframe''.  Would you
call your dog ``dog''?  (Pace Barry Rowlingson! :-) )

cheers,

Rolf Turner


?str may also be useful..


str is indeed useful --- but not here.

R. T.


bests

milton

On Sun, Oct 18, 2009 at 11:10 PM, ANJAN PURKAYASTHA 
anjan.purkayas...@gmail.com wrote:


Hi,
newbie question. I have a data-frame with 3 named columns: Name,  
Obs1,

Obs2.
The Name column members are made of alphanumeric characters: T1,  
T2, T3

etc.
I would like to acess only that subset of the data-frame with Name  
== T44.

X - dataframe[dataframe$Name=='T44'] does not work.
Any ideas on how to do this?
I'm sure I'm missing a simple concept here.

Thanks,
Anjan


--
=
anjan purkayastha, phd
bioinformatics analyst
whitehead institute for biomedical research
nine cambridge center
cambridge, ma 02142

purkayas [at] wi [dot] mit [dot] edu
703.740.6939

   [[alternative HTML version deleted]]

__
R-help@r-project.org mailing list
https://stat.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide
http://www.R-project.org/posting-guide.htmlhttp://www.r- 
project.org/posting-guide.html

and provide commented, minimal, self-contained, reproducible code.



[[alternative HTML version deleted]]

__
R-help@r-project.org mailing list
https://stat.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide http://www.R-project.org/posting- 
guide.html

and provide commented, minimal, self-contained, reproducible code.



##
Attention:\ This e-mail message is privileged and confid...{{dropped:9}}

__
R-help@r-project.org mailing list
https://stat.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide http://www.R-project.org/posting-guide.html
and provide commented, minimal, self-contained, reproducible code.


[R] What is the difference between prcomp and princomp?

2009-10-18 Thread Peng Yu
Some webpage has described prcomp and princomp, but I am still not
quite sure what the major difference between them is. Can they be used
interchangeably?

In help, it says

 'princomp' only handles so-called R-mode PCA, that is feature
 extraction of variables.  If a data matrix is supplied (possibly
 via a formula) it is required that there are at least as many
 units as variables.  For Q-mode PCA use 'prcomp'.


What are R-mode and Q-mode? Are they just too different numerical
methods to compute PCA?

__
R-help@r-project.org mailing list
https://stat.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide http://www.R-project.org/posting-guide.html
and provide commented, minimal, self-contained, reproducible code.


[R] order of interaction coeff in lm

2009-10-18 Thread slurpy

Hi. I haven't found this question asked elsewhere, so I hope I am not missing
something trivial. 

y-rnorm(1:10)
x1-rnorm(1:10)
x2-rnorm(1:10)
x3-rnorm(1:10)
x4-rnorm(1:10)
reg-lm(y~x1*x2+x3+x4)
summary(reg)

The output of this puts x1:x2 after x3 and x4. In my case this is very
cumbersome because I have about 50 dummy variables instead of just x3 and
x4. Is there any way I can get R to print out x1:x2 right after x1 and x2? I
can always do:

x1_x2-x1*x2
reg-lm(y~x1+x2+x1_x2+x3+x4)

But I was wondering if there is something more elegant, since I like the
interaction term denoted by a colon.

Thanks.

R 2.9.2/vista x64
-- 
View this message in context: 
http://www.nabble.com/order-of-interaction-coeff-in-lm-tp25952978p25952978.html
Sent from the R help mailing list archive at Nabble.com.

__
R-help@r-project.org mailing list
https://stat.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide http://www.R-project.org/posting-guide.html
and provide commented, minimal, self-contained, reproducible code.


Re: [R] What is the difference between prcomp and princomp?

2009-10-18 Thread Peng Yu
On Sun, Oct 18, 2009 at 10:42 PM, Peng Yu pengyu...@gmail.com wrote:
 Some webpage has described prcomp and princomp, but I am still not
 quite sure what the major difference between them is. Can they be used
 interchangeably?

 In help, it says

     'princomp' only handles so-called R-mode PCA, that is feature
     extraction of variables.  If a data matrix is supplied (possibly
     via a formula) it is required that there are at least as many
     units as variables.  For Q-mode PCA use 'prcomp'.


 What are R-mode and Q-mode? Are they just too different numerical
 methods to compute PCA?

Also, it seems that 'loadings' of princomp is the same as 'rotation'
of prcomp. I'm wondering whey they have different names in the two
functions.

__
R-help@r-project.org mailing list
https://stat.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide http://www.R-project.org/posting-guide.html
and provide commented, minimal, self-contained, reproducible code.


Re: [R] How to create MULTILEVELS in a dataset??

2009-10-18 Thread Ista Zahn
Hi Saurav,
I was waiting for someone else to answer you, because I'm not sure
I'll be able to explain clearly. But since no one is jumping on it,
I'll take a stab.

On Sun, Oct 18, 2009 at 5:52 PM, saurav pathak pathak.sau...@gmail.com wrote:
 Dear R users

 I have a data set which has five variables. One depenedent variable y, and 4
 Independent variables (education-level, householdincome, countrygdp and
 countrygdpsquare). The first two are data corresponding to the individual
 and the next two coorespond to the country to which the individual belongs
 to. My data set does not make this distinction between individual level and
 country level. Is there a way such that I can make R make countrygdp and
 countrygdpsquare at a different level than the individual level data. In
 other words I wish to transform my dataset such that it recognizes two
 individual level variables to be at Level-1 and the other two country level
 variables at Level-2.


If you're using lmer I don't think you need to do anything special in
terms of data preparation. You will need an explicit country code I
think.

 I need to run a multilevel model, but first I must make my dataset recognise
 data at Level-1 and Level-2. How can I create this country level group (gdp
 and gdp^2) such that I can perform a multilevel model as follows:

 lmer(y ~ education-level + householdincome + countrygdp + countrygdpsquare +
 (1 I Level2),family=binomial(link=probit),data=dataset)

I think you just need to specify country as the grouping variable:

 lmer(y ~ education-level + householdincome + countrygdp +
countrygdpsquare + (1 I
country),family=binomial(link=probit),data=dataset)


 Please kindly help me with the relevant commands for creating this Level2
 (having two variables)

I hope this helps -- I thinks it's less complicated than you were assuming.

-Ista

 Thanks
 Saurav





 Dr.Saurav Pathak
 PhD, Univ.of.Florida
 Mechanical Engineering
 Doctoral Student
 Innovation and Entrepreneurship
 Imperial College Business School
 s.patha...@imperial.ac.uk
 0044-7795321121

        [[alternative HTML version deleted]]

 __
 R-help@r-project.org mailing list
 https://stat.ethz.ch/mailman/listinfo/r-help
 PLEASE do read the posting guide http://www.R-project.org/posting-guide.html
 and provide commented, minimal, self-contained, reproducible code.




-- 
Ista Zahn
Graduate student
University of Rochester
Department of Clinical and Social Psychology
http://yourpsyche.org

__
R-help@r-project.org mailing list
https://stat.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide http://www.R-project.org/posting-guide.html
and provide commented, minimal, self-contained, reproducible code.


Re: [R] Why points() is defined specially for a 1 by 2 matrix?

2009-10-18 Thread Ben Bolker



Peng Yu wrote:
 
 On Sun, Oct 18, 2009 at 10:26 PM, Richard M. Heiberger r...@temple.edu
 wrote:
 points(x[4,],pch=2)# this is plotted as two points

 drops what it sees as an unnecessary dimension.

 Use

 points(x[4,, drop=FALSE], pch=2)

 See FAQ 7.5

 tmp - matrix(1:2)
 tmp
 tmp[,1]
 tmp[,1,drop=FALSE]
 
 Can I specify 'drop' to FALSE by default so that I don't have to
 specify it explicitly?
 
 

 Not that I know of, but things will be easier with the following idiom:

 x = cbind(1:4,3:6)
 plot(x[,1],x[,2],pch=rep(1:2,c(3,1)))

even easier if the plot types correspond to a factor variable in
a data frame:

dat = data.frame(x=1:4,y=3:6,type=factor(c(1,1,1,2)))
with(dat,plot(x,y,pch=as.numeric(type)))

  To answer one of your other questions: ggplot (and lattice) is/are
very powerful, but base graphics are (a) easier to get your head around
and (b) easier to adjust if you don't like the defaults.  Changing things
just a little bit in ggplot can be difficult (as an example, the answer to
your other question about getting rid of grid lines has to do with 
theme_blank(), something like +options(grid.panel.minor=theme_blank())
[try googling theme_blank for a few examples])

-- 
View this message in context: 
http://www.nabble.com/Why-points%28%29-is-defined-specially-for-a-1-by-2-matrix--tp25952698p25953122.html
Sent from the R help mailing list archive at Nabble.com.

__
R-help@r-project.org mailing list
https://stat.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide http://www.R-project.org/posting-guide.html
and provide commented, minimal, self-contained, reproducible code.