.
- Thanks! I expected this also.
nbsp;
Thank you.
nbsp;
-Rommel
nbsp;
nbsp;
nbsp;
nbsp;
- Ursprüngliche Nachricht - Von: Greg Snow-2 [via R] lt;ml-
node+s789695n3838250...@n4.nabble.comgt; Datum: Samstag, 24. September
2011, 12:52 am Betreff: Re: Kolmogorov-Smirnov test
801.408.8111
-Original Message-
From: r-help-boun...@r-project.org [mailto:r-help-bounces@r-
project.org] On Behalf Of Greg Snow
Sent: Monday, September 26, 2011 11:45 AM
To: rommel; r-help@r-project.org
Subject: Re: [R] Kolmogorov-Smirnov test
There are criteria to tell
Are you doing the 2 sample KS test? Comparing if 2 samples come from the same
distribution?
With 3,000 points you will still likely have power to find meaningless
differences, what exactly are you trying to accomplish by doing the comparison?
I am really only familiar with the KS test done in
It looks like Bland-Altman procedures would be appropriate for this project
(http://en.wikipedia.org/wiki/Bland-Altman_plot).
-Original Message-
From: r-help-boun...@r-project.org [mailto:r-help-boun...@r-project.org] On
Behalf Of Pedro Mardones
Sent: Tuesday, September 20, 2011 11:47
Here is one example of doing it, first create a matrix with all your contrasts
(I added some to make it full), then invert it to get the dummy variable
encodings, use the 'contrast' function (or 'C') to set the contrasts, then use
'aov' or 'lm' and those contrasts will be used:
my.contrasts -
Look at the XLConnect package.
--
Gregory (Greg) L. Snow Ph.D.
Statistical Data Center
Intermountain Healthcare
greg.s...@imail.org
801.408.8111
-Original Message-
From: r-help-boun...@r-project.org [mailto:r-help-bounces@r-
project.org] On Behalf Of Marion Wenty
Sent: Friday,
One additional thing to note is that %% will have different precedence than
(something that was pointed out to me based on %% that is in the TeachingDemos
package).
--
Gregory (Greg) L. Snow Ph.D.
Statistical Data Center
Intermountain Healthcare
greg.s...@imail.org
801.408.8111
What null hypothesis are you trying to test? There is a standard null for
linear models that makes sense in a large number of cases, but what the null is
for non-linear regression is not obvious (and the coefficient = 0 may not even
be possibly, let alone interesting). If you can state what
?identify
--
Gregory (Greg) L. Snow Ph.D.
Statistical Data Center
Intermountain Healthcare
greg.s...@imail.org
801.408.8111
-Original Message-
From: r-help-boun...@r-project.org [mailto:r-help-bounces@r-
project.org] On Behalf Of bby2...@columbia.edu
Sent: Thursday, September 15,
It would appear that your file is not a csv file, but we cannot tell for sure
without seeing an example of the data that you are reading in, can you cut and
paste the first few lines of the file? Or post a link to a copy of the file.
If the data is not able to be shared, then create a new
Here is another approach. A linear regression with a single binomial predictor
will give the same results as a pooled t-test (if you insist on non-pooled then
use sapply as previously suggested). The lm function will do multiple
regressions if given a matrix as the y-variable, so you can do a
Setting par(usr=something) does not survive the creating of a new high level
plot. Using par(new=TRUE) is to be avoided if at all possible (it just leads
to problems like yours), it would be better for you to use matlines instead of
matplot which adds lines to the current plot.
If you want to
Are you running a script using 'source' or in batch mode? And do you want the
script to get input from a user instead of from the script? If that is the
case then you need to work a little harder because R assumes that if you are
running in a non-interactive mode (source and batch mode) that
The best approach is to put all those datasets into a list, then you can loop
through the elements of the list, or use lapply or sapply on the list.
If you insist on keeping them outside of a list, or need a loop to put them
into the list, then look at the 'get' function.
-Original
For bounded density estimation look at the logspline package instead of the
regular density function.
--
Gregory (Greg) L. Snow Ph.D.
Statistical Data Center
Intermountain Healthcare
greg.s...@imail.org
801.408.8111
-Original Message-
From: r-help-boun...@r-project.org
The sys.call or match.call functions may be what you are looking for:
tmpfun - function(x,y,z,...) {
+ as.list( sys.call() )
+ }
tmpfun( x=5, w=3, 1:10 )
[[1]]
tmpfun
$x
[1] 5
$w
[1] 3
[[4]]
1:10
tmpfun2 - function(x,y,z,...) {
+ as.list( match.call() )
+ }
tmpfun2( x=5, w=3, 1:10 )
For this example you can work out every possibility, you can use this in place
(or in addition to) the simulations. Here is a quick example with a couple
other ways to look at what is happening:
all.rolls - expand.grid(1:6,1:6)
max.roll - apply(all.rolls, 1, max)
The grconvertX and grconvertY functions may be helpful in finding the endpoints
to use.
--
Gregory (Greg) L. Snow Ph.D.
Statistical Data Center
Intermountain Healthcare
greg.s...@imail.org
801.408.8111
-Original Message-
From: r-help-boun...@r-project.org [mailto:r-help-bounces@r-
If your main goal is to look at a data frame and you are ok with scrolling,
then look at the View function (note capitalization) as an alternative to just
printing the data frame.
--
Gregory (Greg) L. Snow Ph.D.
Statistical Data Center
Intermountain Healthcare
greg.s...@imail.org
801.408.8111
For the initial .Rd file look at the 'prompt' function, it will create a basic
.Rd file based on the current function definition with all the structure in
place, then you go in and edit it to have more meaningful information.
--
Gregory (Greg) L. Snow Ph.D.
Statistical Data Center
The Predict.Plot function in the TeachingDemos package can help you visualize
interactions. It will work best if Month is treated as a continuous variable.
--
Gregory (Greg) L. Snow Ph.D.
Statistical Data Center
Intermountain Healthcare
greg.s...@imail.org
801.408.8111
-Original
Do you want to be doing other things with the R terminal in the meantime?
Are you OK with the terminal being locked up between runs and just want to see
the output updated?
Is it OK to have a new instance of R run the function?
If the last one is doable then you can have your OS scheduled to
If you know how to generate random data that represents your null hypothesis
(chance, auc=0.5) and how to do your analysis, then you can do this by
simulation, simulate a dataset at a given sample size, analyze it, repeat a
bunch of times and see if that sample size is about the right size. If
Here is one approach, whether it is better than the basic loop or not is up to
you:
x - list(A=c(d, e, f), B=c(d, e), C=c(d))
tmp - unlist(x)
tmp2 - sub( '[0-9]+$', '', names(tmp) )
x.new - split( tmp2, tmp )
x.new
$d
[1] A B C
$e
[1] A B
$f
[1] A
Of course this version will have
Use the capture.output function with printing the summary of the anova, then
pass that to your function.
--
Gregory (Greg) L. Snow Ph.D.
Statistical Data Center
Intermountain Healthcare
greg.s...@imail.org
801.408.8111
-Original Message-
From: r-help-boun...@r-project.org
Look at the spread.labs function in the TeachingDemos package, it may do what
you want (see the examples).
--
Gregory (Greg) L. Snow Ph.D.
Statistical Data Center
Intermountain Healthcare
greg.s...@imail.org
801.408.8111
-Original Message-
From: r-help-boun...@r-project.org
If you add column names to your contrast matrix (treat3) then those names will
be used in the coefficient names.
--
Gregory (Greg) L. Snow Ph.D.
Statistical Data Center
Intermountain Healthcare
greg.s...@imail.org
801.408.8111
-Original Message-
From: r-help-boun...@r-project.org
What makes you think that the p-value of 1 is more accurate than the p-value of
0? The K-S test will show significance for very small differences in
distributions when the sample size is big enough.
Also, it is not clear that you are using it correctly. Generally you would
just give the raw
Which package is gofstat in? can you show us your data, or some details about
your data?
Note that the KS test (and all goodness of fit tests) are rule out tests, they
can show that the data is unlikely to come from a distribution, but can never
prove that it does come from a distribution.
Others have shown how to do this, but a better approach may be to load your
.Rdata file into a new environment rather than the global environment, then you
can work with the environment (and all the objects in it) using lapply, or
loops with [[]] instead of fighting with get and assign.
Well there is the esp package, though it is still very pre-alpha. Let's see
what it has to suggest for this question:
source('g:/R/esp/esp.R')
esp()
[1] tabulation overabundant positron carfare contaminant sprawl station swat
aversion
Wow, that almost looks like it should make some sense,
The usual smoothed scatterplot assumes that the x variable (longitude) is fixed
and that the y-variable (latitude) is observed with error, and that the mapping
is 1 to 1 or 1 to many in that for each value of x you can have at most one y
value. These assumptions don't seem to make much sense
?split
--
Gregory (Greg) L. Snow Ph.D.
Statistical Data Center
Intermountain Healthcare
greg.s...@imail.org
801.408.8111
-Original Message-
From: r-help-boun...@r-project.org [mailto:r-help-bounces@r-
project.org] On Behalf Of Jonathan Greenberg
Sent: Thursday, July 28, 2011 2:14
The tkexamp function in the TeachingDemos package can help with creating tcltk
dialog boxes. There are also several other functions in that package that use
tcltk dialogs that you could use as examples to build your own if tkexamp is
not enough for you.
--
Gregory (Greg) L. Snow Ph.D.
I remember seeing an example using the EM algorithm where one of the variables
was age of child and they assumed that an age like 16 months was accurate to
the month, but ages like 18 months may have been off by as much as 2 months and
ages like 3 years could be off by 6 months (or more), so
A small modification of this would be:
library(TeachingDemos)
-3 %=% z %=% 3
Whether that is prettier or uglier than Jim's answer is in the eye of the
beholder (for longer variable names this version could save a few key strokes).
--
Gregory (Greg) L. Snow Ph.D.
Statistical Data Center
Others have explained why R gives a different answer based on a different
approximation, but if you want to get the same answer as the book/minitab/...
for your own understanding (or so the grader doesn't get confused by superior
answers, or other reasons) here is one way to do it:
x - c(
There are several options depending on what exactly you want to do.
If you use code like:
write.table(x, 'clipboard', sep='\t')
In R, then go to excel and choose a cell and paste to it, the matrix or data
frame 'x' will be pasted into excel at that point (topleft corner of data goes
in
, 2011 3:10 PM
To: Greg Snow; jim holtman; Manuel K.
Cc: r-help@r-project.org
Subject: RE: [R] Number in interval
Note that the precendence of %=% is not the
same as that of =, so you can be surprised by its
behavior in slightly more complex expressions:
z - seq(1.2, len=5, by=.7
To add to what David and Duncan wrote: If you want to plot something at a point
where the x coordinate is in user coordinates, but the y-coordinate is
something like the middle of the plot, or 1/5th of the way from the top then
you can use the grconvertY function along with the text function.
Note the warning on the help page for recordPlot. If your colleague is using a
different version of R than you there could be problems.
Another approach (or the start to one that you could build on) is the
plot2script function in the TeachingDemos package (which uses recordPlot
internally).
Try
plot(Well~Raw, subset= Plate==101)
-Original Message-
From: r-help-boun...@r-project.org [mailto:r-help-boun...@r-project.org] On
Behalf Of rstudent
Sent: Friday, July 08, 2011 8:56 AM
To: r-help@r-project.org
Subject: [R] Simple conditional plot
I just started using R last week.
Are your users willing to install an Excel plug-in? If so, look at the RExcel
project, it does what you describe and the common user thinks they are just
using Excel with a plug-in without realizing that they have installed and are
using a useful tool in the background.
-Original
How are you measuring length of stay? A chi-square test suggests that you have
it categorized, a t-test assumes it is continuous (and relatively symmetric
with the amount depending on sample size).
Do you have any censoring? (patients dying or transferring before discharge) if
so you should
In general, if the data frames are all related then it is best to keep them
together in a list like you have. But if you want to change the names of the
component data frames then you can use a loop, or sometimes better use the
lapply function. Here is a basic example:
tmp - list(
You are suffering from the fact that the longest distance between 2 points is a
shortcut.
The df$column notation is a shortcut for df[[column]] that has some nice
properties, but the shortcut gets in the way when you want to do something more
structured. Try qq1[[z]]==y and avoid all that
:31 PM
To: Greg Snow
Cc: Steven Wolf; 'David Winsemius'; r-help@r-project.org
Subject: Re: [R] source, echo...and clicking the mouse
On 30/06/2011 5:33 PM, Greg Snow wrote:
On some operating systems (which we don't know yours, see the posting
guide) the output is buffered and including
In addition to Uwe's answer, you might also want to consider the pairs2
function in the TeachingDemos package. It lets you plot sections of the
overall scatterplot matrix rather than the whole thing, so you could spread the
entire scatterplot matrix over multiple pages.
--
Gregory (Greg) L.
On some operating systems (which we don't know yours, see the posting guide)
the output is buffered and including a call to flush.console() will flush all
the output from the buffer to the console. Put the function call throughout
the script and when it is run it will stop buffering for a bit.
I would look at the ggplot2 package, I have seen examples using ggplot2 that
look similar to what you describe.
-Original Message-
From: r-help-boun...@r-project.org [mailto:r-help-boun...@r-project.org] On
Behalf Of Patrick Jemison
Sent: Monday, June 27, 2011 4:36 PM
To:
Look at the spread.labs function in the TeachingDemos package or the
spread.labels function in the plotrix package.
-Original Message-
From: r-help-boun...@r-project.org [mailto:r-help-boun...@r-project.org] On
Behalf Of David martin
Sent: Friday, June 24, 2011 8:07 AM
To:
Here is one approach (different from capture.output) that I believe
accomplishes what you are trying to do:
frtop - function(){
out - numeric(0)
topenv - environment()
fr - function(x) { ## Rosenbrock Banana function
x1 - x[1]
x2 - x[2]
Try levels(df$city) or unique(df$city) depending on if it is a factor (default)
or character string.
--
Gregory (Greg) L. Snow Ph.D.
Statistical Data Center
Intermountain Healthcare
greg.s...@imail.org
801.408.8111
-Original Message-
From: r-help-boun...@r-project.org
I second the fortune nomination, probably the 1st paragraph is sufficient for a
fortune.
--
Gregory (Greg) L. Snow Ph.D.
Statistical Data Center
Intermountain Healthcare
greg.s...@imail.org
801.408.8111
-Original Message-
From: r-help-boun...@r-project.org [mailto:r-help-bounces@r-
recommend is look through the task view. Maybe someone else has
a better refrence.
--
Gregory (Greg) L. Snow Ph.D.
Statistical Data Center
Intermountain Healthcare
greg.s...@imail.org
801.408.8111
From: Dave Evens [mailto:daveeve...@yahoo.co.uk]
Sent: Friday, June 17, 2011 1:48 AM
To: Greg Snow
Does this do what you want?
x - abs(rnorm(100))
tt - 1:100
m - mean(x)
par(mfrow=c(2,1))
yy - c(0,3)# y-limit
plot(tt,x,type=l,ylim=yy)
abline(h=m)
clip(0,100,0,m)
polygon( c(1,tt,100), c(m,x,m), col='red' )
--
Gregory (Greg) L. Snow Ph.D.
Statistical Data Center
Intermountain Healthcare
What do you want to happen when both are NA? what do you want to happen if both
have values?
--
Gregory (Greg) L. Snow Ph.D.
Statistical Data Center
Intermountain Healthcare
greg.s...@imail.org
801.408.8111
-Original Message-
From: r-help-boun...@r-project.org
I don't think that this approach is appropriate here. Each iteration after the
1st the lm/predict combination will assume that the new data is exact when in
fact it is an estimate with some error involved. To properly do this you need
to take into account that variability. There is a time
You could do this using permutation tests. If everything is balanced and
orthogonal then you can permute the predictors, otherwise you can follow these
main steps:
1. Formulate the test of interest as a full and reduced model test and find the
F statistic (or other, but I will assume F) for
The individual tests on coefficients in logistic regression are generally based
on a Wald test statistic. Unfortunately there is a bit of a paradox possible
in this case where the coefficient is highly significant, but due to a
flattening of the likelihood the standard error is overestimated
It would help us to help you if you told us which package ncredint is in (it is
not in any of the packages that is installed on the computer I am presently
using, but could be in multiple others).
Does this interval match what you are expecting?
library(TeachingDemos)
hpd(qbeta, shape1=4,
This is actually FAQ 7.21. As others have mentioned, the most important part
of the Answer is that it is better to use a list instead.
What searching did you do before posting? Is there some way that the FAQ could
be changed that would have made your searching turn up the FAQ answer?
--
This is actually FAQ 7.21 (though the FAQ does not specify how to create the
name in a loop). As others have mentioned, the most important part of the
Answer is that it is better to use a list instead.
What searching did you do before posting? Is there some way that the FAQ could
be changed
The following link may be of interest. It shows a way to use gsub, but have it
automatically applied while reading the data in rather than converting after:
http://finzi.psych.upenn.edu/Rhelp10/2010-February/229550.html
--
Gregory (Greg) L. Snow Ph.D.
Statistical Data Center
Intermountain
If you are willing to prepend a step then you could:
1. Create an empty plot using your data and type='n' (or just plot the data,
the points will be overwritten), you may want to set the asp argument, or
explicitly do the xlim and ylim arguments.
2. Add the graphic using the rasterImage
This sounds like what is called domains in survey sampling (possibly other
names, but that is what I learned it as). The idea is that you take a random
sample (or the population) then ask a question to determine which domain the
subject is in, then ask the question of interest in the domain of
In the windows GUI you can click on the Help menu and then click on 'FAQ on R'
to see the entire FAQ. Typing help.start() on any OS should open a new browser
window with a link to the FAQ. Going to the R homepage
(http://www.r-project.org) and clicking on the FAQs link on the left will
also
, Greg Snow-2.
here i hav a question, how can i save the value of a list to a vector
or a
matrix? say,
i have a list (data1) and a vector or a matrix (data2).
if i made like this,
data2 - as.vector(data1)
it does not change anything, the data2 is still a list.
what i want to do
this that don't have
the experience to know better.
-Original Message-
From: Prof Brian Ripley [mailto:rip...@stats.ox.ac.uk]
Sent: Friday, June 10, 2011 11:36 PM
To: Greg Snow
Cc: Richard M. Heiberger; r-help
Subject: Re: [R] running R commands asynchronously
On Fri, 10 Jun 2011, Greg Snow wrote
Yes, punif is the function to use, however the KS test (and the others) are
based on an assumption of independence, and if you know that your data points
sum to 1, then they are not independent (and not uniform if there are more than
2). Also note that these tests only rule out distributions
Ph.D.
Statistical Data Center
Intermountain Healthcare
greg.s...@imail.org
801.408.8111
From: kairavibha...@googlemail.com [mailto:kairavibha...@googlemail.com] On
Behalf Of Kairavi Bhakta
Sent: Friday, June 10, 2011 2:16 PM
To: Greg Snow; r-help@r-project.org
Subject: RE: [R] Test if data
Tk windows run asynchronous to the rest of R, so you could write a small
function that lets you type the command into a Tk window and runs it while you
continue to work in the main R, then the window could signal you when the
function finishes.
-Original Message-
From:
Look at the logspline package for one approach to doing this.
-Original Message-
From: r-help-boun...@r-project.org [mailto:r-help-boun...@r-project.org] On
Behalf Of teriri
Sent: Thursday, June 02, 2011 12:19 AM
To: r-help@r-project.org
Subject: [R] [Plea to the R Gods] Theoretical and
The sink function will write to a file what normally shows up on the screen
after running some code. So while it is possible to use it to capture the
output of the mle command and read the results into excel, I don't see anything
useful that you could then do with it in excel.
If you can tell
I did not see any code above, but you could write a simple function that does
the mle fit (is this mle from the stats4 package?) then extracts the
information that you want and puts it into a vector, something like:
out - c( coef(fit), sqrt(diag(vcov(fit))), ll=logLik(fit) )
And returns the
rare
value between 999.99 and 1000.
-Original Message-
From: Bogaso Christofer [mailto:bogaso.christo...@gmail.com]
Sent: Sunday, May 29, 2011 12:15 AM
To: Greg Snow; R-help@r-project.org
Subject: RE: [R] Normality test
Hi Greg, please forgive me as I could not understand one part
To build on Robert's suggestion (which is very good to begin with), you might
consider using the vis.test function in the TeachingDemos package with the
vt.qqnorm function. This will create the qq plot of your data along with
several other qqplots of normal samples of the same size. If you
The msm package may be of use to you.
-Original Message-
From: r-help-boun...@r-project.org [mailto:r-help-boun...@r-project.org] On
Behalf Of phillowe
Sent: Saturday, May 28, 2011 10:51 AM
To: r-help@r-project.org
Subject: [R] Markov Chain model coding
Can anyone help with coding for
There are point in polygon functions in some of the spatial packages that would
tell you if a point is within a polynomial approximation of the ellipse. But
in your case I would take a different approach. Generally confidence ellipses
are based on manhalobis distances, so you can just compute
...@imail.org
801.408.8111
-Original Message-
From: Stephan Kolassa [mailto:stephan.kola...@gmx.de]
Sent: Wednesday, May 25, 2011 3:49 PM
To: Greg Snow
Cc: r-help@r-project.org
Subject: Re: [R] What are the common Standard Statistical methods used
for the analysis of a dataset
Dear all
Data Center
Intermountain Healthcare
greg.s...@imail.org
801.408.8111
-Original Message-
From: Michael Dewey [mailto:i...@aghmed.fsnet.co.uk]
Sent: Wednesday, May 25, 2011 1:14 PM
To: Greg Snow; Ramnath R; r-help@r-project.org
Subject: Re: [R] What are the common Standard Statistical
The only statistical method that I know of that can be applied to any dataset
without further definition of the nature of the data or the question being
asked is SnowsCorrectlySizedButOtherwiseUselessTestOfAnything which is found in
the TeachingDemos package for R. However this test is not
If the x's that don't enter at the same time can be considered independent of
each other, and only clusters that enter at the same time are dependent, then
you can still do a permutation test by creating clusters with dependent values
within each cluster, but independent between clusters, then
).
The random number generators in R are based on good theory, I doubt that there
would be any problems with using the sample function for randomization tests.
From: Wenjin Mao [mailto:wenj@gmail.com]
Sent: Tuesday, May 24, 2011 6:54 PM
To: Greg Snow
Cc: Meyners, Michael; r-help@r-project.org
The EBImage package from bioconductor will read image files (png and others),
the object read has a slot called .Data that is a 3 dimensional array with one
dimension being the color (red,green,blue), so you could just grab one of those
3 layers and it would probably be what you want (or some
From the command line the namespaces may not help much (I am happy to be
corrected). But if I am running a function from the command line then I can
check the search path to see if there is a conflict and make sure to call the
correct one, the bigger problem is when writing another package
I think that the Rcmdr package already does a lot of what you want, rather than
starting from scratch you should check it out. Rcmdr has the functionality of
letting you write your own add-ins, so if it currently does not have the
functions that you want you could just create an add-in to add
Look at the lmList function in the nlme package, it does what I think you want.
--
Gregory (Greg) L. Snow Ph.D.
Statistical Data Center
Intermountain Healthcare
greg.s...@imail.org
801.408.8111
-Original Message-
From: r-help-boun...@r-project.org [mailto:r-help-bounces@r-
If your goal is to end up with a pdf file, then I would suggest creating the
pdf file directly using the pdf function (you can specify height and width in
the function) then run your commands to create the plot and use dev.off() to
finish.
You often get different results when writing directly
Which direction is it truncated? (only values less than a allowed or only
greater?).
One simple approach is rejection sampling, just generate from a regular poisson
distribution, then throw away any values in the truncated region. Another
approach if the legal values are those from 0 to a,
]
Sent: Monday, May 16, 2011 7:46 PM
To: Greg Snow
Cc: r-help@r-project.org
Subject: Re: [R] simulation from truncated poisson
It is truncated from left.
On Mon, May 16, 2011 at 6:33 PM, Greg Snow
greg.s...@imail.orgmailto:greg.s...@imail.org wrote:
Which direction is it truncated? (only values
Look at the help for par, specifically the section on 'mar' to set the per plot
margins smaller and the section on 'oma' to leave room for the overall title.
--
Gregory (Greg) L. Snow Ph.D.
Statistical Data Center
Intermountain Healthcare
greg.s...@imail.org
801.408.8111
-Original
-
From: adele_thomp...@cargill.com [mailto:adele_thomp...@cargill.com]
Sent: Friday, May 13, 2011 2:21 PM
To: Greg Snow; r-help@r-project.org
Subject: RE: [R] Plots: I've deleted axes, now to delete space
Easy fix. Under ?par, I don't see where I can enter an overall title.
Should I add a text
Contrary to the commonly held assumption, the Wilcoxin test does not deal with
medians in general.
There are some specific cases/assumptions where the test/interval would apply
to the median, if I remember correctly the assumptions include that the
population distribution is symmetric and the
One simple way:
Run R (the gui version)
Click on the Edit menu
Click on the GUI Preferences item.
Select the font, size, style, colors, etc. that you want. If you click on Save
then these become the new default. If you click on Apply, but don't save then
they will last that session but not be
Just to add to what David already said, you might want to look at the
Predict.Plot and TkPredict functions in the TeachingDemos package for a simple
interface for visualizing predicted values in regression models.
These plots are much more informative than a single number trying to capture
Look at the dummy.coef function.
--
Gregory (Greg) L. Snow Ph.D.
Statistical Data Center
Intermountain Healthcare
greg.s...@imail.org
801.408.8111
-Original Message-
From: r-help-boun...@r-project.org [mailto:r-help-bounces@r-
project.org] On Behalf Of James Lawrence
Sent:
The strsplit function is probably the closest R function to perls split
function. For more detailed control the gsubfn package can be useful.
-Original Message-
From: r-help-boun...@r-project.org [mailto:r-help-boun...@r-project.org] On
Behalf Of Gamliel Beyderman
Sent: Thursday, May
Will all the keywords always be present in the same order? Or are you looking
for the keywords, but some may be absent or in different orders?
Look into the gsubfn package for some tools that could help.
-Original Message-
From: r-help-boun...@r-project.org
1. ?paste ?sprintf
2. ?par (look at col.axis) ?axis
3. ?pdf ?png ?dev.copy
--
Gregory (Greg) L. Snow Ph.D.
Statistical Data Center
Intermountain Healthcare
greg.s...@imail.org
801.408.8111
-Original Message-
From: r-help-boun...@r-project.org [mailto:r-help-bounces@r-
project.org]
601 - 700 of 2156 matches
Mail list logo