or any consequence.
John Maindonald email: [EMAIL PROTECTED]
phone : +61 2 (6125)3473fax : +61 2(6125)5549
Centre for Mathematics Its Applications, Room 1194,
John Dedman Mathematical Sciences Building (Building 27)
Australian National University, Canberra ACT 0200.
On 17 Aug 2007
.
The SEs and t-statistics that lm() gives for the finally
selected model can be grossly optimistic. Running the
analysis with the same model matrix, but with y-values
that are noise, can give a useful wake-up call.
John Maindonald email: [EMAIL PROTECTED]
phone : +61 2 (6125)3473fax : +61
of insight.
Incidentally, the rlm class inherits from lm, and plot.lm()
(or, preferably, the plot generic) can be used with rlm objects.
This is not the case for lqs objects.
John Maindonald.
On 17 Mar 2007, at 10:00 PM, Andy Liaw wrote:
From: Liaw, Andy [EMAIL PROTECTED]
Date: 17 March 2007
more detail, I will provide it.
It looks a huge improvement. I am sure that R users will be duly
grateful for your efforts.
Regards
John Maindonald.
John Maindonald email: [EMAIL PROTECTED]
phone : +61 2 (6125)3473fax : +61 2(6125)5549
Centre for Mathematics Its Applications
of the
current analysis. Especially in the binary case, this would
sometimes be useful.
John Maindonald email: [EMAIL PROTECTED]
phone : +61 2 (6125)3473fax : +61 2(6125)5549
Centre for Mathematics Its Applications, Room 1194,
John Dedman Mathematical Sciences Building
2nd edn) of Data Analysis Graphics using R, CUP,
2003 and 2006. Install and attach the DAAG package and try
example(moths)
John Maindonald email: [EMAIL PROTECTED]
phone : +61 2 (6125)3473fax : +61 2(6125)5549
Centre for Mathematics Its Applications, Room 1194,
John Dedman
to be 1)
Null deviance: 75.485 on 79 degrees of freedom
Residual deviance: 33.896 on 78 degrees of freedom
AIC: 89.579
Number of Fisher Scoring iterations: 14
John Maindonald email: [EMAIL PROTECTED]
phone : +61 2 (6125)3473fax : +61 2(6125)5549
Centre for Mathematics Its
and is unlikely to do so for a while.
John Maindonald email: [EMAIL PROTECTED]
phone : +61 2 (6125)3473fax : +61 2(6125)5549
Centre for Mathematics Its Applications, Room 1194,
John Dedman Mathematical Sciences Building (Building 27)
Australian National University, Canberra ACT 0200
be larger, or the distributions more dispersed,
than for true maximum likelihood estimates, were you able to
obtain them!
John Maindonald
John Maindonald email: [EMAIL PROTECTED]
phone : +61 2 (6125)3473fax : +61 2(6125)5549
Mathematical Sciences Institute, Room 1194,
John Dedman
money, the issue is not ease of performing regression with huge data
sets, but ease of data exploration.
John Maindonald email: [EMAIL PROTECTED]
phone : +61 2 (6125)3473fax : +61 2(6125)5549
Mathematical Sciences Institute, Room 1194,
John Dedman Mathematical Sciences Building
will have so much memory that memory
management will not be an issue.
John Maindonald email: [EMAIL PROTECTED]
phone : +61 2 (6125)3473fax : +61 2(6125)5549
Mathematical Sciences Institute, Room 1194,
John Dedman Mathematical Sciences Building (Building 27)
Australian National
, following on from (3), I do not understand how lmer() is able to
calculate a t-statistic. There seems to me to be double dipping.
Certainly, I noted a convergence problem.
John Maindonald email: [EMAIL PROTECTED]
phone : +61 2 (6125)3473fax : +61 2(6125)5549
Mathematical Sciences
of the time effects for these
two specific groups.
There is no problem with the time random effect; that can be estimated
from the within group variation in slopes, between subjects.
John Maindonald email: [EMAIL PROTECTED]
phone : +61 2 (6125)3473fax : +61 2(6125)5549
Mathematical
If anyone has a list of application areas where there is
extensive use of R, I'd like to hear of it. My current
short list is:
Bioinformatics
Epidemiology
Geophysics
Agriculture and crop science
John Maindonald
Mathematical Sciences Institute, Australian National University.
[EMAIL PROTECTED
engineering and marketing. (I
hope my summaries are acceptably accurate). I'm not sure what force these
other respondents have given the word extensive.
John Maindonald
Mathematical Sciences Institute
Australian National University.
[EMAIL PROTECTED]
Berton Gunter wrote:
Define extensive.
I
them.
John Maindonald email: [EMAIL PROTECTED]
phone : +61 2 (6125)3473fax : +61 2(6125)5549
Mathematical Sciences Institute, Room 1194,
John Dedman Mathematical Sciences Building (Building 27)
Australian National University, Canberra ACT 0200.
On 4 Jan 2006, at 10:00 PM
of
software, and users reward you by complaining that they want even
more!]
John Maindonald.
On 1 Jan 2006, at 10:00 PM, [EMAIL PROTECTED] wrote:
From: Dave Atkins [EMAIL PROTECTED]
Date: 1 January 2006 1:40:45 AM
To: r-help@stat.math.ethz.ch
Subject: Re: [R] lme X lmer results
Message: 18
Maindonald.
On 31 Dec 2005, at 5:51 AM, Douglas Bates wrote:
On 12/29/05, John Maindonald [EMAIL PROTECTED] wrote:
Surely there is a correct denominator degrees of freedom if the
design
is balanced, as Ronaldo's design seems to be. Assuming that he has
specified the design correctly to lme
pretty well
independently of the lmer results?
John Maindonald.
On 29 Dec 2005, at 10:00 PM, [EMAIL PROTECTED] wrote:
From: Douglas Bates [EMAIL PROTECTED]
Date: 29 December 2005 5:59:07 AM
To: Ronaldo Reis-Jr. [EMAIL PROTECTED]
Cc: R-Help r-help@stat.math.ethz.ch
Subject: Re: [R] lme X lmer
(order = c(0,
John Maindonald email: [EMAIL PROTECTED]
phone : +61 2 (6125)3473fax : +61 2(6125)5549
Centre for Bioinformation Science, Room 1194,
John Dedman Mathematical Sciences Building (Building 27)
Australian National University, Canberra ACT 0200
(arg1, arg2)),
bty=n)
John Maindonald.
On 11 Oct 2005, at 11:41 AM, Spencer Graves wrote:
Have you received a reply to this post? I couldn't find one,
and I couldn't find a solution, even though one must exist. I can
get the substitute to work in main but not legend
As an alternative to the effects package, try predict() with
type=terms
JM
On 7 Oct 2005, at 8:00 PM, Peter Dunn wrote:
From: [EMAIL PROTECTED]
[mailto:[EMAIL PROTECTED] On Behalf Of Peter Dunn
Sent: Wednesday, October 05, 2005 9:06 PM
To: R-help mailing list
Subject: [R] R/S-Plus
element of the legend argument
to legend() becomes, effectively:
substitute(y == a * x^b, list(a = B[1], b=B[2]))
John Maindonald email: [EMAIL PROTECTED]
phone : +61 2 (6125)3473fax : +61 2(6125)5549
Centre for Bioinformation Science, Room 1194,
John Dedman Mathematical Sciences
themselves. As far as I can see, this is
a straightforward use of MCMC to estimate model parameters; it is
not clear to me the results from the lmer() fit are used.
John Maindonald.
On 30 Sep 2005, at 8:00 PM, [EMAIL PROTECTED] wrote:
From: Roel de Jong [EMAIL PROTECTED]
Date: 29 September 2005 11:19
suspect that confusion
between these two contexts in which the word groups is wont to
be used lay behind the use of the EXPNO/COND form of
model formula.
John Maindonald.
On 10 Sep 2005, at 8:00 PM, Larry A Sonna wrote:
From: Larry A Sonna [EMAIL PROTECTED]
Date: 10 September 2005 12:10:06 AM
SPSS only yields 16. Can anyone help?
John Maindonald email: [EMAIL PROTECTED]
phone : +61 2 (6125)3473fax : +61 2(6125)5549
Centre for Bioinformation Science, Room 1194,
John Dedman Mathematical Sciences Building (Building 27)
Australian National University, Canberra ACT 0200
The DAAG package has the following:
show.colors(type=c(singles, shades, grayshades),
order.cols=TRUE)
I am sure there are better ways to do the ordering than my ad hoc
approach,
though.
John Maindonald email: [EMAIL PROTECTED]
phone : +61 2 (6125)3473fax : +61 2(6125
better turns up my
proposal will be to proceed as I have just indicated.
John Maindonald email: [EMAIL PROTECTED]
phone : +61 2 (6125)3473fax : +61 2(6125)5549
Centre for Bioinformation Science, Room 1194,
John Dedman Mathematical Sciences Building (Building 27)
Australian National
is estimated at the level of the random
effect, then of course there are just 12 effects that should
appear in any qq or suchlike plot.
John Maindonald email: [EMAIL PROTECTED]
phone : +61 2 (6125)3473fax : +61 2(6125)5549
Centre for Bioinformation Science, Room 1194,
John Dedman
(voice.part ~ height, data=singer)
print(gph, position=c(0, 0.5, 1, 1)) # x0, y0, x1, y1
par(fig=c(0, 1, 0,0.5), new=TRUE) # x0, x1, y0, y1
boxplot(height ~ voice.part, data=singer, horiz=TRUE)
John Maindonald email: [EMAIL PROTECTED]
phone : +61 2 (6125)3473fax : +61 2(6125)5549
), but it is probably not the result that you want!
See further pp.320-321 of the DAAG book.
John Maindonald email: [EMAIL PROTECTED]
phone : +61 2 (6125)3473fax : +61 2(6125)5549
Centre for Bioinformation Science, Room 1194,
John Dedman Mathematical Sciences Building (Building 27)
Australian
records
nfields
3 4
6 1
tab - table(nfields)
(1:length(nfields))[nfields == 4]
[1] 2
readLines(oneBadRow.txt, n=-1)[2]
[1] 11 13 1 6
Note the various option settings for count.fields()
John Maindonald email: [EMAIL PROTECTED]
phone : +61 2 (6125)3473fax : +61 2(6125)5549
Centre
combination has little effect the
variability may be large, while if it has a dramatic effect
(kills everything!), there may be no variability to speak of.
John Maindonald.
On 22 Feb 2005, at 10:06 PM, [EMAIL PROTECTED] wrote:
To: 'Bob Wheeler' [EMAIL PROTECTED]
Cc: r-help@stat.math.ethz.ch
Subject
) (is that strictly correct?)
I often want to use xtable() with tables. There is no method
defined for the class table. After a bit of rummaging, I found
that I can use:
xtable(as(tab, matrix)).
John Maindonald email: [EMAIL PROTECTED]
phone : +61 2 (6125)3473fax : +61 2(6125)5549
Centre
Guide. This can be a bit
hard to find. With limma installed, type help.start().
After some time a browser window should open. Click on
Packages | limma | Overview | LIMMA User's Guide (pdf)
John Maindonald email: [EMAIL PROTECTED]
phone : +61 2 (6125)3473fax : +61 2(6125)5549
panel.superpose() in this way?
The new abilities provided by trellis.focus() etc add
greatly to the flexibility of what can be done with lattice
plots. The grid-lattice combination is a great piece of
software.
John Maindonald email: [EMAIL PROTECTED]
phone : +61 2 (6125)3473fax : +61 2(6125
the newpage=FALSE for the next print()
is ignored. Should I be able to do something like this?
Clearly I do not understand what happens when
trellis.focus() is invoked.
This seems an area where an effective GUI, with a
graphical display of the viewport tree, could be very
helpful.
John Maindonald
.
John Maindonald email: [EMAIL PROTECTED]
phone : +61 2 (6125)3473fax : +61 2(6125)5549
Centre for Bioinformation Science, Room 1194,
John Dedman Mathematical Sciences Building (Building 27)
Australian National University, Canberra ACT 0200
), the GUI/noGUI issue
will be a minor consideration, and hours or weeks spent
learning R will be at most a modest consideration.
John Maindonald email: [EMAIL PROTECTED]
phone : +61 2 (6125)3473fax : +61 2(6125)5549
Centre for Bioinformation Science, Room 1194,
John Dedman
a scatterplot.
[cf MDSplot() for randomForest()]
John Maindonald email: [EMAIL PROTECTED]
phone : +61 2 (6125)3473fax : +61 2(6125)5549
Centre for Bioinformation Science, Room 1194,
John Dedman Mathematical Sciences Building (Building 27)
Australian National University, Canberra ACT 0200.
On 5
The output has, mostly, not been revised. Thus it will in
some cases reflect an earlier version of R. This is a task
for some later time.
John Maindonald email: [EMAIL PROTECTED]
phone : +61 2 (6125)3473fax : +61 2(6125)5549
Centre for Bioinformation Science, Room 1194,
John Dedman
This pretty much uses regular latex, with the page dimensions
changed and the font attributes redefined to make them
larger than usual inside the slide environment. Be sure to
load the packages xspace and colortbl as well as pdfscreen.
John Maindonald email: [EMAIL PROTECTED]
phone : +61
for
the types of problems that typically come up in those areas.
John Maindonald.
John Maindonald email: [EMAIL PROTECTED]
phone : +61 2 (6125)3473fax : +61 2(6125)5549
Centre for Bioinformation Science, Room 1194,
John Dedman Mathematical Sciences Building (Building 27)
Australian
of aov1 and aov2 (where the
errors are indeed crossed) in lme4.
John Maindonald.
On 12 Aug 2004, at 8:03 PM, [EMAIL PROTECTED] wrote:
I will follow the suggestion of John Maindonald and present the
problem by example with some data.
I also follow the advice to use mean scores, somewhat reluctantly
on the likely
impact. Use of simulate.lme() seems like a good idea.
John Maindonald.
On 11 Aug 2004, at 8:05 PM, [EMAIL PROTECTED] wrote:
From: Spencer Graves [EMAIL PROTECTED]
Date: 10 August 2004 8:44:20 PM
To: Gijs Plomp [EMAIL PROTECTED]
Cc: [EMAIL PROTECTED]
Subject: Re: [R] Enduring LME confusion
handle some
standard requirements.
John Maindonald.
On 30 Jun 2004, at 8:02 PM, Barry Rowlingson
[EMAIL PROTECTED] wrote:
A colleague is receiving some data from another person. That person
reads the data in SAS and it takes 30s and uses 64k RAM. That person
then tries to read the data in R
to explain (to myself and others)
what happens, to say that there is both a lexical stack and a call
stack. Is that a legitimate use of terminology?
John Maindonald email: [EMAIL PROTECTED]
phone : +61 2 (6125)3473fax : +61 2(6125)5549
Centre for Bioinformation Science, Room 1194
a great deal of
practically oriented comment (e.g., wrt variable selection)
that I have found useful. There is useful critical comment
on inappropriate use of p-values in such contexts.]
John Maindonald.
On 3 May 2004, at 9:58 AM, Frank E Harrell Jr wrote:
I'm sorry to have taken so long
to speculate how R's model fitting routines
might be tuned to allow a Bayesian interpretation. What
family or families of priors would be on offer, and/or used by
default? What default mechanisms would be suitable
useful for indicating the sensitivity of results to the choice
of prior?
John Maindonald
somewhere in the undergrowth?
John Maindonald.
Frank Harrell wrote:
They [p-values] are objective only in the sense that
subjectivity is deferred in a difficult to document way
when P-values are translated into decisions.
The statement that frequentist methods are the norm, which I'm
afraid is usually
exercises
(the more challenging exercises) on the web, via a link
from http://wwwmaths.anu.edu.au/~johnm/r-book.html
Anyone who would find this inconvenient should contact
me directly.
I will shortly post a provisional list on the above web site.
John Maindonald email: [EMAIL PROTECTED
be estimated. Notice that
the BLUPs are pulled back towards the overall mean, relative to
the group means.
NB also, specify level=1 to incorporate the random group (Rail)
effects into the predicted values.
John Maindonald email: [EMAIL PROTECTED]
phone : +61 2 (6125)3473fax : +61 2(6125
models work.
__
[EMAIL PROTECTED] mailing list
http://www.stat.math.ethz.ch/mailman/listinfo/r-help
John Maindonald email: [EMAIL PROTECTED]
phone : +61 2 (6125)3473fax : +61 2(6125)5549
Centre for Bioinformation Science, Room 1194,
John
.
The very small P[positive test | no infection] is an incomplete
part of the story.
Regrettably I suspect that the result would be much the same for
AIDS or genetics counsellors anywhere.
John Maindonald email: [EMAIL PROTECTED]
phone : +61 2 (6125)3473fax : +61 2(6125)5549
Centre
54 matches
Mail list logo