try:
example(TukeyHSD)
hth,
Kingsford Jones
On Thu, Feb 19, 2009 at 9:51 AM, Saeed Ahmadi ahmadi_sa...@yahoo.com wrote:
Hi,
I have a basic and simple question on how to code pairwise (multiple) mean
compariosn between levels of a factor using one of the Duncan, Tukey or LSD.
Thanks
before we see such memory in laptops:
http://www.ubergizmo.com/15/archives/2009/01/samsung_opens_door_to_32gb_ram_stick.html
Like you, I'd also be interested in hearing about configurations folks
have used to work w/ large datasets.
hth,
Kingsford Jones
On Mon, Feb 16, 2009 at 5:10 AM
On Thu, Feb 12, 2009 at 3:12 PM, Johannes Huesing johan...@huesing.name wrote:
Last time I tried, rseek.org yielded no results when searching for inferno.
...although, if you hit the 'Support Lists' tab it finds the thread in
which Patrick announced it.
--
Johannes Hüsing
is printed vs what is stored
iv) see
http://cran.r-project.org/doc/FAQ/R-FAQ.html#Why-doesn_0027t-R-think-these-numbers-are-equal_003f
v) rather than subtracting from 1, use the 'lower.tail' argument to pt:
2*pt(10.843,2838,lower=F)
[1] 7.185635e-27
hth,
Kingsford Jones
G-lm(y~x1+x2+x3+x4+x5
or simultaneously estimate the coefficients and variance structure via
nlme::gls and its 'weights' argument...
On Tue, Feb 10, 2009 at 7:57 AM, John Fox j...@mcmaster.ca wrote:
Dear Kishore,
Yes, White's heteroscedasticity-consistent standard errors are just that --
standard errors for the
See ?[ and its examples
Also, section 2.7 of An Introduction to R is a good place to start:
http://cran.r-project.org/doc/manuals/R-intro.html#Index-vectors
hth,
Kingsford Jones
On Mon, Feb 9, 2009 at 5:27 PM, jjh21 jjhar...@gmail.com wrote:
Hello,
I am trying to do some data cleaning
Also, see this page for more ideas on mapping an auxiliary variable by county:
http://r-spatial.sourceforge.net/gallery/
(e.g. plots 9, 14, 15 and 21)
Kingsford Jones
On Fri, Feb 6, 2009 at 3:22 PM, Greg Snow greg.s...@imail.org wrote:
Googling for Denmark Shapefile leads to this website
,] 1.000 0.9963557 -0.9883690 0.8548065
[2,] 0.9963557 1.000 -0.9976663 0.8084090
[3,] -0.9883690 -0.9976663 1.000 -0.7663847
[4,] 0.8548065 0.8084090 -0.7663847 1.000
kappa(x)
[1] 2813.326
hth,
Kingsford Jones
On Thu, Jan 29, 2009 at 7:00 PM, Joseph P Gray jpg...@uwm.edu wrote
On the topic of visualizing correlation, see also
Murdoch, D.J. and Chow, E.D. (1996). A graphical display of large
correlation matrices.
The American Statistician 50, 178-180.
with examples here:
# install.packages('ellipse')
example(plotcorr, package='ellipse')
On Sat, Mar 8, 2008 at 3:01
see
?subset
Or use indexing, which is covered in section 2.7 of an introduction to
R (but note that a data frame has 2 dimensions)
hth,
Kingsford Jones
On Sun, Jan 25, 2009 at 3:06 PM, pfc_ivan pfc_i...@hotmail.com wrote:
I am a beginner using this R software and have a quick question.
I
How about:
a - c(1,2,3,3,2,1,6,3,2)
b - c(NA,a[-length(a)])
c - c(a[-1],NA)
a[b==1 c==3]
[1] 2 6
hth,
Kingsford Jones
On Sun, Jan 25, 2009 at 3:02 PM, Jörg Groß jo...@licht-malerei.de wrote:
Hi,
I have a quit abstract problem, hope someone can help me here.
I have a vector like
with predict function or it's some bug in my
code?
...perhaps the levels of 'sample' differ between 'traindata' and 'testdata'?
hth,
Kingsford Jones
Thanks much for your help!
Chun
Chun Zhang
Statistician at Roche
__
R-help@r-project.org mailing list
try
install.packages('car')
?car::linear.hypothesis
hth,
Kingsford Jones
On Wed, Jan 21, 2009 at 4:20 PM, Roberto Patuelli
roberto.patue...@lu.unisi.ch wrote:
Dear All,
I was wondering if it is possible to generate a regression summary (it does
not matter at this stage if from an lm
Unangu,
If you haven't seen the 200pg workshop notes that Adrian Baddeley has
made available from his spatstat webpage, I highly recommend them:
http://www.csiro.au/files/files/pn0y.pdf
hth,
Kingsford Jones
On Sat, Jan 10, 2009 at 9:04 PM, Unangu una...@gmail.com wrote:
To understand some
I hope that Marc doesn't mind, but I felt that part of his recent post
was important enough to deserve it's own subject line rather then
being lost in a 60-msg-long thread...
On Sun, Jan 11, 2009 at 10:08 AM, Marc Schwartz
marc_schwa...@comcast.net wrote:
...
I strongly believe that the
accept work done in R, and
b) SAS is the only way to effectively handle data?
best,
Kingsford Jones
On 7 Jan, 14:50, Marc Schwartz marc_schwa...@comcast.net wrote:
on 01/07/2009 08:44 AM Kevin E. Thorpe wrote:
Zaslavsky, Alan M. wrote:
This article is accompanied by nice pictures
did you install and load the 'ctv' package?
install.packages('ctv')
library('ctv')
then try your code...
Kingsford Jones
On Sat, Jan 10, 2009 at 1:00 PM, oscar linares wins...@gmail.com wrote:
Dear Rxperts,
Using R 2.8.1 and trying
install.views(Cluster)
getting error
Error: could
,]6 36216
[11,] 91 8281 753571
[12,] 41 1681 68921
[13,] 22 484 10648
[14,]39 27
best,
Kingsford Jones
So
I think the code below it, although not too short, does what the person
asked. Thanks though because I understand outer better now.
temp - matrix(c
software that can help illuminate data trends more clearly
blah blah blah blah
;-)
Kingsford Jones
http://www.flickr.com/photos/sluggerotoole/153603564/
Simon.
--
Simon Blomberg, BSc (Hons), PhD, MAppStat.
Lecturer and Consultant Statistician
Faculty of Biological and Chemical Sciences
$version.string
[1] R version 2.8.0 (2008-10-20)
Kingsford Jones
On Mon, Jan 5, 2009 at 10:18 AM, William Dunlap wdun...@tibco.com wrote:
Arg, the 'sapply(...)' in the function was in the initial
comment,
gm - function(x, group){ # medians by group:
sapply(split(x,group),median)
but someone's
126.630.42 129.21
gm
user system elapsed
37.560.84 38.78
sapply/gm user ratio: 3.371406
On Mon, Jan 5, 2009 at 2:37 PM, Kingsford Jones
kingsfordjo...@gmail.com wrote:
Here's some more timing's of Bill's function. Although in this
example sapply has a clear performance advantage
rather
than analysis. R in combination with a GIS is really the way to go.
Here is a current book that covers many of the spatial tools available in R
http://www.springer.com/public+health/epidemiology/book/978-0-387-78170-9
hope that helps,
Kingsford Jones
this kind of study?
Thank you
resending to provide a more informative subject line
On Fri, Jan 2, 2009 at 3:21 PM, Kingsford Jones
kingsfordjo...@gmail.com wrote:
Hi David,
A general answer to your question is: yes, R would be useful for such
analyses - particularly when interfaced with a GIS. For an
introduction
repeated sampling the
fitted line will vary less than the observed y values at any given set
of x-values, and this is reflected in the fact that the confidence
interval is narrower than the prediction interval.
hth,
Kingsford Jones
On Wed, Dec 31, 2008 at 2:15 PM, Dylan Beaudette
dylan.beaude
(4,8,4,2))
plot(confint(mult))
hth,
Kingsford Jones
On Mon, Dec 8, 2008 at 5:06 PM, Metconnection [EMAIL PROTECTED] wrote:
Dear R-users,
I'm currently using the multcomp package to produce plots of means with 95%
confidence intervals
i.e.
mult-glht(lm(response~treatment, data=statdata
[48,] D E C
[49,] E A B
[50,] E A C
[51,] E A D
[52,] E B A
[53,] E B C
[54,] E B D
[55,] E C A
[56,] E C B
[57,] E C D
[58,] E D A
[59,] E D B
[60,] E D C
Kingsford Jones
On Tue, Dec 2, 2008 at 9:41 PM, G. Jay Kerns [EMAIL PROTECTED] wrote:
Dear Brandon,
On Tue, Dec
see
?coef # extract the estimates
?vcov # extract their covariance matrix
?qf # get the F quantile of interest
Also, you may be interested in
?car::ellipse
?ellipse::ellipse.lm
?gmodels::glht.test
hth,
Kingsford Jones
On Sat, Nov 29, 2008 at 4:30 PM, Kyle Matoba [EMAIL PROTECTED] wrote
[students$Sex == 1])
hth,
Kingsford Jones
On Sun, Nov 30, 2008 at 2:15 PM, Simone Gabbriellini
[EMAIL PROTECTED] wrote:
sorry for my bad presentation...
read.spss gives me this:
students
$Auno
[1] 6 1 2 2 1 3 4 2 4 2 4 4 1 1 NA 1 4 2 1 1 1 5 4
[24] 2 1 2 1 2 1
weights because you didn't specify any through the
weights argument to lm. If you had, you could extract them using the
same technique used in the above code: w - mod.fit$weights
hth,
Kingsford Jones
Thanks for your time and kind
help!
Dana
Steven McKinney wrote:
Hi Dana,
Many
include working with
unbalanced data, not producing negative variance estimates when the
MS_{error} is larger than the MS_{between groups}, and providing a
great deal of flexibility in structuring both the random effects and
error covariance matrices.
hth,
Kingsford Jones
On Thu, Nov 13, 2008 at 7
RSiteSearch('vec vech', restrict = 'fun') turns up several packages
with vec and vech functions.
E.g., matrixcalc, fUtilities, ks, ...
hth,
Kingsford Jones
On Thu, Oct 23, 2008 at 11:47 AM, megh [EMAIL PROTECTED] wrote:
Can anyone please tell whether there is any R function to act as VEC
-level factor and covar is quantitative
confint(fit, level = .90)
hth,
Kingsford Jones
I suspect that several participants in this board have implemented
code to so this. I hope that someone is willing to share the code.
Thanks in advance.
Dennis
Dennis Fisher MD
P (The P Less Than
your system. Thank You.
---BeginMessage---
RSiteSearch('vec vech', restrict = 'fun') turns up several packages
with vec and vech functions.
E.g., matrixcalc, fUtilities, ks, ...
hth,
Kingsford Jones
On Thu, Oct 23, 2008 at 11:47 AM, megh [EMAIL PROTECTED] wrote:
Can anyone please tell whether
your system. Thank You.
---BeginMessage---
RSiteSearch('vec vech', restrict = 'fun') turns up several packages
with vec and vech functions.
E.g., matrixcalc, fUtilities, ks, ...
hth,
Kingsford Jones
On Thu, Oct 23, 2008 at 11:47 AM, megh [EMAIL PROTECTED] wrote:
Can anyone please tell whether
in two
means using an ANCOVA perhaps something like...
fit - lm(y ~ trt + covar) # trt is a 2-level factor and covar is quantitative
confint(fit, level = .90)
hth,
Kingsford Jones
I suspect that several participants in this board have implemented
code to so this. I hope that someone is willing
in two
means using an ANCOVA perhaps something like...
fit - lm(y ~ trt + covar) # trt is a 2-level factor and covar is quantitative
confint(fit, level = .90)
hth,
Kingsford Jones
I suspect that several participants in this board have implemented
code to so this. I hope that someone is willing
package (although it does not appear to
be on CRAN).
hth,
Kingsford Jones
Regards a lot.
Savava.
Miltinho Astronauta
Brazil
On Wed, Oct 8, 2008 at 6:02 AM, Clément Calenge
[EMAIL PROTECTED] wrote:
It's very kind of Stephen to plug my book, but it's notwhat you're
looking
You may find this site useful:
http://lme4.r-forge.r-project.org/bib/lme4bib.html
On Tue, Oct 7, 2008 at 9:08 AM, Dieter Menne
[EMAIL PROTECTED] wrote:
Julia S. julia.schroeder at gmail.com writes:
Now, I did that in my article and I got a response from a reviewer that I
additionally should
to be useful (formulas for crossed random
effects are much more straightforward than in nlme). If not, this
page provides some helpful tips on eliciting useful responses from
this list: http://www.r-project.org/posting-guide.html
HTH,
Kingsford Jones
On Sat, Sep 27, 2008 at 2:21 AM, Gregor
is looking for the 'typeof' page.
Or even
example(type)
Also, after a brief introduction, the R Language Definition document
begins with a discussion of types.
Kingsford Jones
It would have been better to have called the function now known as
``is.integer()''
something like ``is.storedAsInteger
Bo,
Try using RSiteSearch with the strings 'huber-white', 'sandwich' or
even 'clustered standard errors'.
You may also want to consider a mixed models approach -- see:
http://www.stat.columbia.edu/~cook/movabletype/archives/2007/11/clustered_stand.html
HTH,
Kingsford Jones
On Tue, Sep 16
On Wed, Sep 17, 2008 at 3:11 PM, Gillian Silver [EMAIL PROTECTED] wrote:
What would I do if I have something like:
x - rnorm(1:1000)
y - rnorm(1:1000)
z - x + y
and I want the rainbow to increase with z? (i.e., red for lowest z...all the
way up to the last color in the rainbow for the
On second thought, this is more likely to be what you're looking for...
library(rgl)
x - rnorm(1000)
y - rnorm(1000)
z - x + y
plot3d(x, y, z, col = rainbow(1000, end = 5/6)[rank(z)], size = 3)
On Wed, Sep 17, 2008 at 4:06 PM, Kingsford Jones
[EMAIL PROTECTED] wrote:
On Wed, Sep 17, 2008 at 3
Keeping Gavin's advice in mind, you may also want to look at ?acf (and
see section 14.1 of MASS) and help(ACF, package=nlme) (see section 5.3
of MEMSS). These are useful functions for exploring the 1d empirical
autocorrelation structure of model residuals.
hth,
Kingsford Jones
On Fri, Aug 15
and V7 is about .991
Kingsford Jones
On Thu, May 15, 2008 at 3:01 PM, e-letter [EMAIL PROTECTED] wrote:
Thank you to all you eagle eyes; amendment made accordingly and solved. Not
sure how the difference occurred...
[[alternative HTML version deleted
that someone more 'mathy'
than me could explain. Tools like lme, lmer, proc mixed and proc
glimmix fall into this category.
hoping this helps,
Kingsford Jones
Thanks again.
cheers,
Rolf
function, which might be
the right tool for your analyses (
https://stat.ethz.ch/mailman/listinfo/r-sig-mixed-models ).
Kingsford Jones
Cheers,
Federico
--
Federico C. F. Calboli
Department of Epidemiology and Public Health
Imperial College, St Mary's Campus
Norfolk Place, London W2 1PG
on this list that gls is underused (
http://tolstoy.newcastle.edu.au/R/help/06/08/32487.html ). Given the
fact that it can drastically reduce the constraints of the constant
variance and independence assumptions of the ubiquitous linear model,
I agree.
good luck,
Kingsford Jones
Discussion on weights
possible ways to get prediction intervals:
https://stat.ethz.ch/pipermail/r-sig-mixed-models/2008q2/thread.html#841
Finally, why assume a Poisson error distribution for a binary response?
Kingsford Jones
On Wed, May 7, 2008 at 10:13 AM, Bert Gunter [EMAIL PROTECTED] wrote:
Sorry, my reply below may
see ?library
You need to load the package before R can access its functions.
Kingsford Jones
On Mon, May 5, 2008 at 12:18 PM, [EMAIL PROTECTED] wrote:
Hello,
I'm a new user of the R environment. I need to do some power analysis. For
this purpose, I installed the pwr package from the R
, but if you had counts
rather than presence/absence it could be.
best,
Kingsford Jones
On Fri, May 2, 2008 at 2:27 PM, Christian Hof [EMAIL PROTECTED] wrote:
Dear all,
how can I, with R, transform a presence-absence (0/1) matrix of species
occurrences into a presence-only table (3 columns
the Spatial Task View:
http://cran.r-project.org/web/packages/rgdal/index.html
Also, there is a r-sig for spatial data:
https://stat.ethz.ch/mailman/listinfo/r-sig-geo
Kingsford Jones
On Thu, May 1, 2008 at 6:56 AM, stephen sefick [EMAIL PROTECTED] wrote:
Does anyone know of a package to make a map
I sent the wrong link for the spatial Task View. Here it is
http://cran.r-project.org/web/views/Spatial.html
On Thu, May 1, 2008 at 8:21 AM, Kingsford Jones
[EMAIL PROTECTED] wrote:
There are many ways to do this, depending on what type of spatial data
you are working with and what you want
to hang
around here a while. I don't know of a nice collection (but I'm sure there
are several.)
Duncan Murdoch
;-) here's one: https://stat.ethz.ch/pipermail/r-help/2005-October/080991.html
Kingsford Jones
__
R-help@r-project.org mailing list
that the usual specification for varPower is
varPower(form = ~ your.formula), but by default the mean is used. See
Ch 5 of the Pinheiro and Bates Mixed-effects Models book for details)
Kingsford Jones
__
R-help@r-project.org mailing list
https://stat.ethz.ch
finding
the help of someone who is an expert in these types of models (known
as mixed, hierarchical, multilevel, random effects, variance
components, nested ANOVA, etc).
best,
Kingsford Jones
Thanks
__
R-help@r-project.org mailing list
in the documents then the onus is on me to try to get my ideas
incorporated (see
http://wiki.r-project.org/rwiki/doku.php?id=misc:rpatch ).
Kingsford Jones
On Thu, Apr 24, 2008 at 9:37 AM, Duncan Murdoch [EMAIL PROTECTED] wrote:
On 4/24/2008 12:08 PM, Martin Maechler wrote:
Hmm,
KeBe == Beck
101 - 157 of 157 matches
Mail list logo