For points 4 and 5, you could use a robust linear fit. One way to do that
is to use rlm() from package MASS, which is used in several examples in
the book that package MASS supports.
On Sun, 4 May 2008, Tobias Erik Reiners wrote:
Dear Helpers,
I just started working with R and I'm a bit ove
G'day Phil,
On Sun, 4 May 2008 14:05:09 +1000
phil colbourn <[EMAIL PROTECTED]> wrote:
> rlnorm takes two 'shaping' parameters: meanlog and sdlog.
>
> meanlog would appear from the documentation to be the log of the
> mean. eg if the desired mean is 1 then meanlog=0.
These to parameters are the
rlnorm takes two 'shaping' parameters: meanlog and sdlog.
meanlog would appear from the documentation to be the log of the mean.
eg if the desired mean is 1 then meanlog=0.
So to generate random values that fit a lognormal distribution I would
do this:
rlnorm(N , meanlog = log(mean) , sdlog
Tobi,
I think that it would be easier to provide advice if you were more
explicit on what the model will be used for, and what is the structure
of the data. Is there only one measurement for each marsupial? Is
the goal to
a) produce a model to predict marsupial weight given other variables,
and
Dear Helpers,
I just started working with R and I'm a bit overloaded with information.
My data is from marsupials reindroduced in a area. I have weight(wt),
hind foot
lenghts(pes) as continues variables and origin and gender as categorial.
condition is just the residuals i took from the model
Hi, Mike:
Thanks for the bug report. Attached please find a version of
'curfit.free.knot' that will now fit splines of degree other than 3.
This will appear in the next official release of DierckxSpline. Until
then, you can use the attached. (It still has known bugs, but with luck
th
In the case of muitivariate, from the documentation it looks like I can
compare more than two signals at a time.
Each column of the input matix seem to accommodate a signal.
The problem is that my signals do NOT have the same number of samples
(length).
They were all collected at 30Hz so the sampli
Thanks very much...it is exactly what I needed, and I'm a bit embarassed
that I couldn't find it on my own.
One might consider adding "aggregate" to the "See also:" lines of by and
tapply. That would have prevented me from needing to email the list (which I
may have accidentally done twice; I apo
Dear Colleagues,
Apologies for a long email to ask what I feel may be a very simple
question; I figure it's better to overspecify my situation.
I was asked a question, recently, by a colleague in my department
about pre-aggregating variables, i.e., computing the mean of defined s
Hello all
Please ignore this last post as I now realize it is because the stacked
components are summed, as they should be. I needed the columns to be
stacked but not summed so just plotted the two series separately with add=T
Sorry...
Murray Richardson
Original Message
This should do what you want -- not sure what happens after the 4th less than 5.
> a <- rep(c(3,5,7), 4)
> b <- rep(c(NA,1,2), 4)
> df <- data.frame(a,b)
> # determine which ones are < 5 and count the occurances
> less.5 <- cumsum(df$a < 5)
> # create new value
> df$c <- ifelse(df$a < 5, (less.5 +
On 03/05/2008 3:29 PM, Murray Richardson wrote:
Hello users,
I've noticed a problem when creating a stacked column plot when a column
contains a negative and a positive value. e.g.
series1<-c(-1,-2, 3, 4, 5)
series2<-c( 5, -4,-3,-2, 1)
data<-rbind(series1,series2)
barplot(as.matrix(data), be
Hello, R-helpers.
I have a data frame below called df. I want to add a variable c, based on
the following logic:
if df$a < 5 then the first two times this condition is met place a 1 in c,
the next two times this condition is met place a 2 in c and when the
condition is not met let c equal df$b.
Hello users,
I've noticed a problem when creating a stacked column plot when a column
contains a negative and a positive value. e.g.
series1<-c(-1,-2, 3, 4, 5)
series2<-c( 5, -4,-3,-2, 1)
data<-rbind(series1,series2)
barplot(as.matrix(data), beside=FALSE)
In these cases (i.e. first, third a
Dear Deepayan,
I am sorry to bother you again, but I have some more questions on the R
wireframe :
- when plotting an interaction between two variables, originating from a
logistic model with several factors : what happens to the other variables (ie
the ones not plotted)? Are they set to 0? O
user parameter "offset"
barplot(d,offset = -2) #assuming that d is your data
Summer Nitely wrote:
>
> Hi,
> I have data that is on a log base 2 scale. It goes from negative factors
> of
> 2 to positive ones. I am using barplot. However, I don't want the data
> centered at 0 - I want the min
"Nelson Castillo" <[EMAIL PROTECTED]> wrote in
news:[EMAIL PROTECTED]:
> On Sun, Apr 27, 2008 at 10:10 AM, David Winsemius
> <[EMAIL PROTECTED]> wrote:
>> "Nelson Castillo" <[EMAIL PROTECTED]> wrote in
>> news:[EMAIL PROTECTED]:
>
snip
>
> That was exactly what I needed :-) Thanks a lot. I tri
R-help is not at all the place to ask such a question.
Please use ess-help (or an Xemacs mailing list).
> "WS" == Werner Stanzl <[EMAIL PROTECTED]>
> on Fri, 2 May 2008 20:41:21 -0400 writes:
WS> Hi,
WS> When I start Xemacs version 21.5.28 (in Unix), using the init.el:
Hi List,
I'm following on the discussions regarding the use of exploratory factor
analysis with binary data, from nabble:
http://www.nabble.com/PCA--for-Binary-data-to11050371.html#a11092127
and i tried using FactorMineR package but unfortunately i have trouble
understanding what are the typ
Mark, Henrique, and Jorge,
Thank you all very much for your replies. I'm really impressed by your
knowledge of R: I'd never even heard of of some of these functions
(especially split() and unlist().)
I went with the Henrique-Mark solution and it is working fine. Thanks
again!
David
[[
Hi, we're kicking off a new predictive analytics online training program and
allowing people to sign up at half price until May 21st. The online training
starts July 7th and is on demand and self-paced. It covers about 1/3 of our
in-person training seminar, and is targeted to managers and markete
Mark,
I concur. It has helped me very much in the short time I have used it.
Very concise and well written.
Thanks Phil!
[EMAIL PROTECTED] wrote:
Since we're on the topic of book reviews, I just received Phil
Spector's new R book called "Data Manipulation with R" and it is also
quite a
ni
I'd rescale my data to minimum around 0 and use some separate axis()
that takes the rescaling into account - given there is no package that
already does it. ;-)
Uwe Ligges
Summer Nitely wrote:
Hi,
I have data that is on a log base 2 scale. It goes from negative factors of
2 to positive one
Karen Liu wrote:
Hello:
I am using plot() in rpart, making a decision tree plot. I use plot()
first, then text() followed, but I think due to the figure vs. text
size, the bottom of the text always gets cut off. I can only "see" the
text of the bottom of the tree if I make the font size RE
24 matches
Mail list logo