I meant fitting not maximising, it is a nonlinear mixed effects
model, with both fixed and random effects. My assumption is that for
the function I am using the approximation approach used in nlme is
not quite close enough, and nothing much that I can do, except for
looking at starting
Hi all,
I'm trying to rank a couple of factors by a variable and a weight of
the variable in each occurrence (some samples are bigger than others).
input = data.frame(
alfa = rnorm(5000),
weight = rnorm(5000,-5,10),
tag1 = sample(c(a,b,c,d),5000,replace=TRUE),
tag2 =
2005/12/18, Bart Joosen [EMAIL PROTECTED]:
Hi,
I have a problem with fitting a model:
I made a dataframe with this data:
a - 1:3
b - 1:3
c - c(3, 2, 3, 2, 1, 2, 3, 2, 3)
df - expand.grid(a,b)
df$result - c
names(df) - c(A,B, result)
Although I can make a graph
Hello, I am a beginner with R and I would need some help with doing barplots.
My problem is that I would like to include both diffrent colors of the bars
and precence/absence of shading lines in the barplots. When reading in the
help file about the col command it states:
col: a vector of
Dear all,
I run the example of censboot contained in boot package. But, I can't
find the confidence interval of the resulted censboot object. Any idea ?
aml.fun - function(data) {
+ surv - survfit(Surv(time, cens)~group, data=data)
+ out - NULL
+ st - 1
+
Hi all,
Before the amount of data given has grown i was initially using read.table to
load the values inside R.
It was feeding my needs because i could tell read.table h=T, then use attach to
access the values by columns names.
Now it takes 20 seconds to load the data's and the first
On Sö, 2005-12-18, 17:32, David STADELMANN skrev:
Hi,
I am running glm logit regressions with R and I would like to test a
linear combination of coefficients (H0: beta1=beta2 against H1:
beta1beta2). Is there a package for such a test or how can I perform
it otherwise (perhaps with logLik()
I am trying to smooth a dataset with evenly spaced values of x,
perhaps using loess smoothing or something similar. However, the y
values are hypergeometrically distributed; I think I want to use a
logarithmic link function. It falls under the general heading of
non-parametric regression. The
On Mon, 2005-12-19 at 12:27 +0100, Björn Rogell wrote:
Hello, I am a beginner with R and I would need some help with doing
barplots.
My problem is that I would like to include both diffrent colors of the
bars
and precence/absence of shading lines in the barplots. When reading
in the
help
read.table _does_ use scan, so why do you claim scan is faster?
There is so much you have not told us that it is impossible to know what
you want. Please do read the posting guide and its references and try to
ask a question that is not predicated on a falsehood.
On Mon, 19 Dec 2005, [EMAIL
DeaR comRades:
I have a 2D spatial binomial process as shown in the data and code below.
I am plotting the number of trials and the number of successes in the spatial
binomial experiments and would like to draw the spatial cells were the trials
and successes were counted, i.e. a partial grid in
Hi Spencer.
When using 'optim' and the first try fails you could:
1) try some other methods: Nelder-Mead, BFGS, ...
2) increase the maximum number of iterations (argument maxit in the control
list)
3) specify the argument parscale in the control list, in order to have all
parameters of same
On Mon, 2005-12-19 at 11:17 -0200, Ruben Roa wrote:
DeaR comRades:
I have a 2D spatial binomial process as shown in the data and code below.
I am plotting the number of trials and the number of successes in the spatial
binomial experiments and would like to draw the spatial cells were the
I have a vector of dates.
I wish to find the month end date for each.
Any suggestions?
e.g.
For 12/15/05, I want 12/31/05,
For 10/15/1995, I want 10/31/1995, etc
__
[[alternative HTML version deleted]]
Here is one way using POSIX: (you can create a function to do this)
x - as.POSIXlt('2005-12-16') # a date
x
[1] 2005-12-16
dput(x) #structure of the date
structure(list(sec = 0, min = 0, hour = 0, mday = 16, mon = 11,
year = 105, wday = 5, yday = 349, isdst = 0), .Names = c(sec,
min,
Sometimes its just one parameter that's the culprit so just use
a grid to get the starting value. Since its known
that in logistic growth the saturation level is a problem we conjecture
that in this one m is the culprit and grid over it. Note that with this
approach we did not need to extend the
Forgot you were asking for the end date, so just subtract a day:
seq(x, by='month', length=2)[2] - 24*3600
[1] 2005-12-31 EST
On 12/19/05, jim holtman [EMAIL PROTECTED] wrote:
Here is one way using POSIX: (you can create a function to do this)
x - as.POSIXlt('2005-12-16') # a date
x
Are you familiar with Pinheiro and Bates (2000) Mixed-Effects Models
in S and S-Plus (Springer)? I suspect that book, especially the latter
half, might contain the information you seek.
spencer graves
p.s. PLEASE do read the posting guide!
The zoo package has a yearmon class with as methods which can be
used:
library(zoo)
dd - Sys.Date() # test data
as.Date(as.yearmon(dd), frac = 1)
as.yearmon converts the Date class date to a year and month of
class yearmon dropping the day and representing it internally in
a way consistent
Or add a month, then subtract a day:
Ndays - function(posix.ct.dates,days) {
# one day = 60*60*24 = 86400 seconds
ans - as.POSIXct(posix.ct.dates) + 86400*days
# we only have a problem if the date went from
# DST to ST or from ST to DST
ans + (as.POSIXlt(posix.ct.dates)$isdst
Spencer:
(warning: highly biased, personal opinions)
My $.02
Looking now at your output, I notice that Corr between
(Intercept) and trust.cz1 for the Random Effects commid is
1.000. This says that the structure of your data are not adequate to
allow you to distinguish between
On Mon, 19 Dec 2005, Marc Schwartz (via MN) wrote:
On Mon, 2005-12-19 at 11:17 -0200, Ruben Roa wrote:
DeaR comRades:
I have a 2D spatial binomial process as shown in the data and code below.
I am plotting the number of trials and the number of successes in the
spatial
binomial
If you work much with time data and time series data and you have not
already mastered the zoo package, you may be interested in a small
testimonial from one unfamiliar with the names of Achim Zeileis and
Gabor Grothendieck two years ago: The package itself seems to have many
useful
Douglas Bates wrote:
The Laplace method in lmer and the default method in glmm.admb,
which according to the documentation is the Laplace approximation,
produce essentially the same model fit. One difference is the
reported value of the log-likelihood, which we should cross-check, and
another
Hi,
aggregate() does not preserve the order of levels for
ordered factors, e.g.,
levs - c(Low, Med, Hi)
d - data.frame(x = 1:30, fac = ordered(rep(levs, 10), levels = levs))
out - aggregate(d[,x], by = list(fac=d$f), FUN = mean)
cat(Original ordered levels:, levels(d$fac), \n)
Dear R-help: I am trying to re-produce the simulations
of Kosorok and Lin (1999) JASA, (The versatility of function-indexed
weighted log-rank test, 320-332).
I need help to generate random variables in R
with hazard function:
h(t) = 2*t*exp{ 0.96 * exp(-4*t^2) }
for t 0.
This is the
Hi!
I have made a guide describing how to install Rcmdr and JGR on SUSE 10.0.
Here it is:
http://www.hellmund.dk/RSUSEGUI.html
If you have installed R on SUSE with the package found on CRAN - and
all the packages on my list was present at the time of installation of
R - it might not be necessary
On 12/19/05, Hans Julius Skaug [EMAIL PROTECTED] wrote:
Douglas Bates wrote:
The Laplace method in lmer and the default method in glmm.admb,
which according to the documentation is the Laplace approximation,
produce essentially the same model fit. One difference is the
reported value of
I'm not certain I understand your question. However, I did the
following as I thought it might produce something that you could use:
RSiteSearch(vector field correlation)
This produced 25 hits, several of which might interest you. In
particular, I wonder of you
I have seen no replies to this post, and I don't know that I can
help, either. However, I wonder if you tried RSiteSearch with your
favorite key words and phrases? For example, I just got 107 hits for
'RSiteSearch(wavelets)'. I wonder if any of them might help you.
If
If the y values are hypergeometrically distributed then they are counts,
right? Loess is designed for continuous, reasonably symmetric data, and so
is inappropriate. You should probably consider GLM for a parametric fit; or
perhaps GAM for a nonparametric fit. As the data appear to have the
Hi,
Vikram and I are beginning work on a native R package to interface
into X-12-arima.We have looked through gretl, and previous
discussions on kludgy interfaces involving calls to the x12a binary.
Our aim is to port as much of the functionality as possible with
native R
This doesn't look like an R question, as I know of no pre-packaged
functionality publicly available that can fit the model that Elizabeth
described, and it doesn't seem like she's particularly interested in an
R-based answer, either.
My gut feeling is that if there is a test of significance for
Hi,
I have written a user defined function that carries
out a monte carlo simulation and outputs various
stats. I would like to access and store the simulation
data (a two column local variable) from outside the
function I have written. How can I output the data
from the function as new variable,
34 matches
Mail list logo
