thanks for the help up to here.
A little problem remains.
I have different ABNR, if i try it with another ABNR, the Column extend
for each ABNR, it should start with FIRST again.
dat1 - read.table(text=FS_ID ABNR
932733688812
11391 33688812
11392 33688812
11388 33688812
11390 33688812
Dear All,
Does anyone know if there is any way to obtain the variance-covariance
matrix for any arbitrarily given estimates with the glm() function?
Here is what I really want. Given an arbitrary estimate (i.e., as starting
points with the start= argument), the glm() function could return the
What you are really asking is how environment variables get set outside of R,
which is very OS-specific and not on topic here, and there can be numerous ways
for it to happen even on a single OS. I will warn you that the path syntax can
vary in Windows depending on where it gets set so beware
Hi Dila,
Try:
transform(melt(dat,id.var=Day),Month=match(variable,month.abb),Year=2012,Amount=value)[,-c(2:3)]
A.K.
On Tuesday, February 25, 2014 11:41 PM, dila radi dilarad...@gmail.com wrote:
Dear Arun,
sorry for asking again.
Assume that we used the sama data set
dat -
Hello all
My upcoming project involves getting the live data from the Team Foundation
Server. I goggled it how to get the data from TFS server via R-Script. But
could not find any satisfactory result.
If anyone has worked with the similar scenario, can you please give me a head
start or else
I have one issue at the moment while I am using a function, fitdistr in
package MASS.
I have 12 years of wind data with 30 min average interval.
My data look like below:
Timewindspeed
1/01/2005 0:00 15
1/01/2005 0:30 11
1/01/2005 1:00 11
1/01/2005 1:30 13
1/01/2005 2:00 15
The robust variances are a completely different estimate of standard error. For linear
models the robust variance has been rediscovered many times and so has lots of names: the
White estimate in economics, the Horvitz-Thompson in surveys, working independence
esitmate in GEE models,
Ok. That's weird. If the stage= variables are indicators for levels of a
factor their cph standard errors look a bit small for 102 events.
Could you show what coxph() gives? It has both the model-based and robust
standard errors. I've just tried an example with fairly extreme weights and
Hi Mat,
Please check the str() of your dat2.
If I change
dat2$DATE - as.POSIXlt(dat2$DATE)
dcast(ddply(merge(dat1,dat2,by=FS_ID)[,-1],.(ABNR), mutate,
DATE=seq_along(DATE)),ABNR~DATE,value.var=POST)
#Error in attributes(out) - attributes(col) :
# 'names' attribute [9] must be the same
Hi,
I have been running successfully Westfall procedures to adjust p-values
in multiple comparisons.
However, the last model I tried consumes too much memory (40 Gb).
Is there a way to reduce memory requirements? Or a workaround to run
multiple comparison tests with Westfall procedures?
Hi Mat,
Please check the str() of your dat2.
If I change
dat2$DATE - as.POSIXlt(dat2$DATE)
dcast(ddply(merge(dat1,dat2,by=FS_ID)[,-1],.(ABNR), mutate,
DATE=seq_along(DATE)),ABNR~DATE,value.var=POST)
#Error in attributes(out) - attributes(col) :
# 'names' attribute [9] must be the same
Hi,
Check the ?str() of my.table.
library(XML)
u='http://www.ininternet.org/calorie.htm'
tables1 = readHTMLTable(u)
my.table1=tables1[[9]]
with(my.table1,mean(PROTEINE))
#[1] NA
#Warning message:
#In mean.default(PROTEINE) :
# argument is not numeric or logical: returning NA
str(my.table1)
This is mostly a guess...
At the end of January, your script succeeded.
Your OS was built Sun Feb 23 05:14:37 PST 2014.
Your script failed on Feb 26.
You probably had not made any changes to R between the end of January and
Feb 26 (I don't see any mention of having done so, and R does not update
Many thanks Rolf,
These codes below are ok :
LinearModel.1 - lm(GDP.per.head ~ Competitivness.score + Quality.score,
data=Dataset)
summary(LinearModel.1)
predict(LinearModel.1, se.fit = FALSE, scale = NULL, df = Inf,interval =
c(prediction),level = 0.95, type = c(response))
One problem
I nominate the following for the fortunes package.
. I still don't think it would be fair to the data, and you
don't want those data liberation people parading around your laboratory
with pictures of helpless data being devoured by a Babbage Difference
Engine.
--
Gregory (Greg) L. Snow
This will probably depend greatly on which operating system you are
using and which progress bar function(s) you are using, neither of
which is clear from your message. If you tell us what your OS is (and
possibly how you are running R, GUI, commandline, etc.) and show us
some sample code
R-helpers:
I was curious if anyone developed a package/approach to installing
packages directly from the R-forge SVN subsystem (rather than waiting
for it to build)? I can, of course, SVN it command line but I was
hoping for an install.packages(svn://) sort of approach. Cheers!
--j
--
Hello,
is it possible to add a primary key to a sqlite table with Rsqlite?
Without create a complex query string?
I have tried dbWriteTable but don't found an option to define a field as
primary key.
Has someone suggestions?
Thanks,
Christian
__
Thanks Jeff.
What I actually ended up doing is closing down RStudio and deleting the other
folder where I didn't want it to store and re-installing those packages from
the deleted folders. So far so good.
Dan
-Original Message-
From: Jeff Newmiller [mailto:jdnew...@dcn.davis.ca.us]
Hello R-users,
I am a biginner in R and I would like to fit a sinusoidal curve to my data. I
couldn't get a nice results.
here is an extract of my data:
line time meters
1 1 04:273.1
2 2 10:480.9
3 3 16:492.9
4 4 23:001.0
5 5 05:033.1
6 6 11:29
I am running Windows 7 64bit. Running a script on Revolution R or R Studio
(the behavior is the same.) The sample script reproduces the problem:
pb - winProgressBar(test progress bar, Some information in %,
0, 100, 50)
Sys.sleep(0.5)
u - c(0, sort(runif(20, 0, 100)), 100)
I understand that the robust variances may lead to a different
standard error. I want the standard error valid for heteroscedastic
data, ultimately, because I have very good estimates of the
measurement variances (why I'm doing weighted fits in the first
place).
For the simple example here, the
Imagine you were plotting two sets, (x1, y1) and (x2, y2) on the same
plot, where x1 and x2 have the same range, but y1 and y2 need different
axis scales.
Here's an example that I think illustrates the idea. It's a long way from
being general, and if I were making it into a function in a package,
Hi. Does anyone know of a package that can take the three attributes,
longitude, latitude, and heatvalue and generate a heat map?
Thank you.
[[alternative HTML version deleted]]
__
R-help@r-project.org mailing list
Well, ?image and ?heatmap (! what else?) can do this -- except that
you probably want this on some sort of map projection, right? Assuming
that's the case, check the spatial task view on CRAN and/or post on
the r-sig-geo list . I am sure there are many packages that can do
various flavors of this.
On 27/02/14 05:43, varin sacha wrote:
Many thanks Rolf,
These codes below are ok :
LinearModel.1 - lm(GDP.per.head ~ Competitivness.score +
Quality.score, data=Dataset)
summary(LinearModel.1)
predict(LinearModel.1, se.fit = FALSE, scale = NULL, df = Inf,interval =
c(prediction),level = 0.95,
When in doubt, first consult the included online help.
?predict.lm
Offers this example, which seems to meet the request
x - rnorm(15)
y - x + rnorm(15)
predict(lm(y ~ x))
new - data.frame(x = seq(-3, 3, 0.5))
predict(lm(y ~ x), new, se.fit = TRUE)
pred.w.plim - predict(lm(y ~ x), new,
On Feb 26, 2014, at 12:24 AM, Xiaogang Su wrote:
Dear All,
Does anyone know if there is any way to obtain the variance-covariance
matrix for any arbitrarily given estimates with the glm() function?
Here is what I really want. Given an arbitrary estimate (i.e., as starting
points with
Dear list,
My Problem Setting Looks as Follows:
I have a list of object [O1,O2,O3,O4,..O6000], on which a pairwise
distance matrix is defined, called dist.matrix.
After that, a agnes has been conducted on the distance matrix.
agn1-agnes(dist.matrix,method=ward)
even cutted
Did you look at the help for agnes.object as directed by the help file
that you yourself linked in your email?
If none of those components meet your needs, as.hclust() - also
pointed to by the agnes.object helpfile - might give you what you
want.
Sarah
On Wed, Feb 26, 2014 at 3:08 PM, Tianchen
Thanks David for stopping by this thread. I have to admit that the word
arbitrary was arbitrarily chosen. As I explained in the second email in
the thread, what I am really interested are some estimates obtained via
other methods that have the same asymptotic distribution as MLE. The
following
I ran the example code in the winProgressBar help file and it ran in normal
display mode (not minimized).
Have you run your code in RGui? Those other tools may be affecting
winProgressBar.
---
Jeff Newmiller
Hi Burnette,
Try:
indx - grep(\\bMW\\b|\\bOT\\b|\\bposter\\b|\\brecords\\b, rownames(res3))
indx1 - grep(Grand Total,rownames(res3))
res3R
-
as.data.frame(matrix(NA,ncol=ncol(res3),nrow=nrow(res3),dimnames=
list(rownames(res3),c(rate of Count of Case ID, rate of Sum of
BW_Due, rate of Sum
Sorry, I have never used these functions before. There is the msgWindow
function, but it works on a graphic device number, and may not work on the
external pointer object that winProgressBar returns.
---
Jeff Newmiller
Sorry that I forgot to include R Help in the addressee list. Here is one of
my earlier follow-up emails.
-- Forwarded message --
From: Xiaogang Su xiaogan...@gmail.com
Date: Wed, Feb 26, 2014 at 8:48 AM
Subject: Re: [R] Fitting glm with maxit=0?
To: Prof Brian Ripley
Here is another one. Thanks all again for the help!
-- Forwarded message --
From: Xiaogang Su xiaogan...@gmail.com
Date: Wed, Feb 26, 2014 at 9:38 AM
Subject: Re: [R] Fitting glm with maxit=0?
To: Thomas Lumley tlum...@uw.edu
Thanks much Prof. Lumley for the tip. That'll be
I have a dataframe that looks like the below. I want to convert the
Captured.Time field to a date object. but some of the entries are not
properly formated and I get a message saying
Error in as.POSIXlt.character(as.character(x), ...) :
character string is not in a standard unambiguous format.
Hi,
I have 20,000 time series variants whose length is about 3000 , and i need to
calculate the mean value of each variant. I invoke R with a java client named
RCaller, and I assigned the values of these time series variants using RCaller.
The problem is : After I start running, the java
Hello,
Could you provide an example of unproperly formatted entry?
Regards,
Pascal
On 27 February 2014 15:03, Bill william...@gmail.com wrote:
I have a dataframe that looks like the below. I want to convert the
Captured.Time field to a date object. but some of the entries are not
properly
Hi. Thanks. Actually I don't know which ones are causing problems. I cannot
search through it because it is quite large (15 million records)
On Thu, Feb 27, 2014 at 7:22 PM, Pascal Oettli kri...@ymail.com wrote:
Hello,
Could you provide an example of unproperly formatted entry?
Regards,
If all you want is the mean, why not do it in Java? That sounds like a half
gigabyte of memory in the best case, and you are trying to move it around from
process to process, making copies here and there. If you are on 32 bit OS then
that will fragment your memory pretty quick.
Anyway, you
Hello,
Did you tried the strptime function?
Regards,
Pascal
On 27 February 2014 15:24, Bill william...@gmail.com wrote:
Hi. Thanks. Actually I don't know which ones are causing problems. I cannot
search through it because it is quite large (15 million records)
On Thu, Feb 27, 2014 at 7:22
I just tried this:
z - strptime(radSampTrim$Captured.Time[1:30], %d%b%Y)
z[1:3]
[1] NA NA NA
On Thu, Feb 27, 2014 at 7:39 PM, Pascal Oettli kri...@ymail.com wrote:
strptime
[[alternative HTML version deleted]]
__
R-help@r-project.org
Hello together,
i have a data.frame like this one:
IDONE TWO THREE
1 2 57
2 6 NA NA
3 5 7NA
4 1 NA NA
Now i want a new column with the max-Value of ONE, TWO and THREE. The
result look like this one:
IDONE TWO
Hello,
I think this should be
z - strptime(radSampTrim$Captured.Time[1:30], %F %T)
Regards,
Pascal
On 27 February 2014 15:58, Bill william...@gmail.com wrote:
I just tried this:
z - strptime(radSampTrim$Captured.Time[1:30], %d%b%Y)
z[1:3]
[1] NA NA NA
On Thu, Feb 27, 2014 at
Yes! that worked.
By the way, still puzzled about how to write the function that would delete
ones where there was an error in the function I used. Any idea on that?
On Thu, Feb 27, 2014 at 8:06 PM, Pascal Oettli kri...@ymail.com wrote:
z - strptime(radSampTrim$Captured.Time[1:30], %F %T)
Hello,
Maybe one possiblity is to search complete cases, such as
complete.cases(z), and subset according to the output.
It might also be interesting to check why you sometimes get NA.
Regards,
Pascal
On 27 February 2014 16:11, Bill william...@gmail.com wrote:
Yes! that worked.
By the way,
47 matches
Mail list logo