Fernando,
I don't have time to do all that you asked, but here is some code that makes
violin plots with mean, median, and 95% CI. I like this plot very much, even
if boxplot purists think it is horrible :-)
I think the boxplot was developed before we had computing power. Now we can
show the det
That doesn't make sense. Which version of Tinn-R are you using?
>From Options->Main->Application
do you get a dialog box with a tab "R", and within that a tab "General"?
At the bottom of the "General" tab there is a button under "Rgui". That
should bring up a file selection box. Proceed to the
Look at this disscussion from two weeks ago:
http://www.nabble.com/adding-the-mean-and-standard-deviation-to-boxplots-td15271398.html
Chad Junkermeier wrote:
>
> I really like the ease of use with the boxplot command in R. I would
> rather have a boxplot that shows the average value and t
Hadley makes a good point. Boxplots should be standardized. Thre is already
inconsistency in what value the "wiskers" represent.
Violin plots with means and 95% CI are a good option if you want to show the
shape of a distribution with the mean. You could add the median as well.
That's what I did
I ran into the same problem when trying to launch R from MS-Access.
The solution was to create a DOS batch file and execute it through
VBA with the shell command. The batch file contains just one line:
C:\pathname\r\bin\rterm.exe --no-save < C:\pathname\rcode.r
HTH
Duncan Murdoch-2 wrote:
I can't determine what is going on in you example, but my approach would be
to read the text file into a text editor that will display hidden characters
(tab, etc.) so you can see the pattern. It could be that there is an extra
tab in some locations. You could then use the editor's replace functi
Harsh-7 wrote:
>
> Hi R users,
>
> I'd be interested in what R users think about social networking around all
> things R. For this, I've set up a social network @
> www.rstuff.socialgo.comand it would be great if you could post your
> comments on the forum created
> for this discussion.
>
>
What I think you are missing is that you didn't change ddd. The ifelse
statement does not assign values to the ddd object. To change ddd it would
read:
ddd <- ifelse ( ddd>360, ddd-360, ddd )
So when you enter "print(ddd)" you get the content of the original object,
which has not changed.
Wha
( ddd>360, ddd-360, ddd )
adding the print function to the ifelse in your first example, executed
using source(), will give you this output:
[1] "numeric"
[1] 461 213 238 249 251
[1] 101 213 238 249 251
[1] 461 213 238 249 251
>
HBaize wrote:
>
>
> What I think yo
Thank you Philippe.
That is very helpful.
Philippe Grosjean wrote:
>
> Hello Rusers,
>
> I have worked on a R Wiki page for solutions in exporting R graphs,
> especially, the often-asked questions:
> - How can I export R graphs in vectorized format (EMF) for inclusion in
> MS Word or Op
It works great on my Acer Aspire One AOD150-1165 10.1 inch, under Windows
XP. I opted to expand to two gigs of RAM because R loads all objects into
active memory. The expansion of 2 gigs over the 1 gig that was standard only
cost $20. Unless you are using large data sets with extensive computatio
## Create directory with unique name based on values of x and y
dir.create(paste(PathName, x,"_",y,sep=""))
Yasin Hajizadeh wrote:
>
> Hi all
>
> I have 2 loops with 2 counters, say outer loop counter is X and inner
> loop counter is Y. In outer loop a data file with known number of
Josh,
Check into "add=TRUE" :-)
All you need to do is insert "add=TRUE" to the second
chart to superimpose it on the first chart. You might also
consider making the second bar a different shade of color
rather than using shadding lines, then it would only be one
plot.
data(HairEyeColor)
a
Well perhaps we can take the mystery of why it work for me, but not you, off
line :-)
I'm using R 2.6.1 on a WinXP platform.
You could try just using shades of color like this:
data(HairEyeColor)
a <- as.table( apply(HairEyeColor, c(1,2), sum) )
a1<-a[1:2,]
barplot(a1,
type="n",
You are drawing four box plots, not two. Two of them are just the number 5.
The two box plots that are only "5" don't have a box, so you can't see that
they're red. Try this:
boxplot(1:19,20:39,col=c("red","blue"))
Paul Adams-8 wrote:
>
> Hello everyone,
>
>
> I am trying to color two boxpl
The problem is dat is a data object, not a function. You used the syntax for
a function "dat(1:19)"
What you probably want is:
gc <- dat[,1:19]
act <- dat[,20:39]
That will select columns 1 through 19 and put them into the object gc, and
act will get columns 20 through 39. Is that what you wa
I have been using RODBC to connect to an Access database to capture data to
create plots. Recently I found incomplete charts. Upon investigation I
discovered that the data retrieved stopped at 3276 rows (records) out of a
table with over 5600 records.
I've tried changing "max","buffsize", and "
I think the function describe() in the package "psych" will give you
want you want. There are other similar functions in the library "Simple"
as well.
Harold
nmarti wrote:
>
> I'm looking for a function that lists a few summary stats for a column (or
> row) of data. I'm aware of summary(x
Could that be extended to generate a population data set with known skew and
kurtosis?
If so, how?
Thanks in advance for suggested solutions.
Harold
Daniel Nordlund-2 wrote:
>
>
>
> Something like this may help get you started.
>
> std.pop <- function(x,mu,stdev){
> ((x-mean(x))/sd(x
Luca,
I ran your code using SPSS 17.0 and R 2.8.1.
I found that the correct date was returned in R using this code:
test.df$newdate <- as.Date(as.POSIXct(test.df$mydate , origin="1582-10-14"))
Also note that SPSS 17.0 is case sensitive in storing variable names, so the
R code in your post need
I think most people find it odd at first if they have always used "=" but
quickly you get use to it and nothing could be more clear. It is explicit.
It is active and provides a direction, a value goes into an object. The
equal sign for assignment is ambiguous.
As an example
x = 3
we only kn
There are many ways to do it. The following will place a blue point on the
boxplot at the mean, then print the mean at the bottom of the plot. In some
plots I've gone too far and included median points and values as well. You
could also put 95% CI on the same plot, but it would get perhaps too "bu
22 matches
Mail list logo
