Displaying 20 results from an estimated 900 matches similar to: "svd"
2010 Aug 15
2
band pass filter
Hello list,
Is there any way to bandpass filter in R
thanks
nuncio
--
Nuncio.M
Research Scientist
National Center for Antarctic and Ocean research
Head land Sada
Vasco da Gamma
Goa-403804
[[alternative HTML version deleted]]
2010 Aug 02
1
removing spatial auto correlation
Hi list,
I am trying to fit arima model for a grid of 360x161x338 points,
where 360x161 is the spatial dimension and 338 is the number of time steps I
have, which is seasonal. For this purpose I used the auto.arima function in
forecast package. After fitting residuals at each grid in space, the auto
correlations are still significant ( but < 0.2). This make me think that the
data
2010 Jul 05
2
timeseries
Dear useRs,
I am trying to construct a time series using as.ts function, surprisingly
when I plot
the data the x axis do not show the time in years, however if I use
ts(data), time in years are shown in the
x axis. Why such difference in the results of both the commands
Thanks
nuncio
--
Nuncio.M
Research Scientist
National Center for Antarctic and Ocean research
Head land Sada
Vasco da Gamma
2010 Jun 07
1
prewhiten
HI all.,
I have some univariate time series that need to be prewhitened. HOw this can
be performed in R.
I am thinking of to fit an ARIMA model and substract this from the original
series. Is this the correct way
THanks in advance
nuncio
--
Nuncio.M
Research Scientist
National Center for Antarctic and Ocean research
Head land Sada
Vasco da Gamma
Goa-403804
[[alternative HTML version deleted]]
2010 May 31
1
missing values in autocorelation
Hi all,
I am trying to find the autocorrelation of some time series. I
have say 100 files, some files have only missing values(-99.99, say). I dont
want to exclude these files as they represent some points in a grid. But
when the acf command is issued i get an error.
Error in plot.window(...) : need finite 'ylim' values
In addition: Warning messages:
1: In min(x) : no
2011 Mar 17
1
Extracting columns from a class
Hi list,
I am not a frequent user of R. Recently I used R in principal
component analysis and got the result as a class, which has information like
standard deviation and principal components from 1 to 10. How is it
possible to extract the column corresponding to first principal component
and write it to a file
the out from prcomp command is something like this
Standard
deviations:
2010 Jul 06
1
acf
Hi list,
I have the following code to compute the acf of a time series
acfresid <- acf(residfit), where residfit is the series
when I type acfresid at the prompt the follwoing is displayed
Autocorrelations of series ?residfit?, by lag
0.0000 0.0833 0.1667 0.2500 0.3333 0.4167 0.5000 0.5833 0.6667 0.7500 0.8333
1.000 -0.015 0.010 0.099 0.048 -0.014 -0.039 -0.019 0.040 0.018
2011 Jun 13
1
documentation in R
How we can call auto.arima in R.
Is there any cran package we need to install for this function?
--
Siddharth Arun,
4th Year Undergraduate student
Industrial Engineering and Management,
IIT Kharagpur
[[alternative HTML version deleted]]
2011 Apr 01
1
principal components
HI all,
I am trying to compute the EOF of a matrix using prcomp but unable to get
the expansion co-efficients.
is it possible using prcomp or are there any other methods
thanks
nuncio
--
Nuncio.M
Research Scientist
National Center for Antarctic and Ocean research
Head land Sada
Vasco da Gamma
Goa-403804
[[alternative HTML version deleted]]
2010 Jul 22
1
tsdiag
HI list,
I want to know whether tsdiag uses k-(p+q) as the lag in ljung box
test. How is it possible to save those values
nuncio
--
Nuncio.M
Research Scientist
National Center for Antarctic and Ocean research
Head land Sada
Vasco da Gamma
Goa-403804
[[alternative HTML version deleted]]
2010 May 20
2
writing autocorrelation and partial auto correlation functions to a file
Dear All,
I am very new to T. I need to fit a ARIMA model to my time
series. So I found the auto correlation functions and partial auto
correlation function in R. Now I want to save these valuse along with the
significance levels to a file. How to do that?. I tried some function in R
like write.table but returns an error "cannot coerce class "acf" into a
2010 Jul 23
1
sink function
I have the following code to write the output from auto.arima function. The
issue is not in finding the model but to divert its out put
fit to a file order_fit.txt. code runs but nothing is written to
order_fit.txt
where am I going wrong
library(forecast)
for (i in 1:2) {
filen = paste("file",i,".txt",sep="")
data <- read.table(filen)
dat1 <- data[,1]
xt <-
2010 Jul 23
0
auto.arima
HI list,
I am using auto.arima from forecast package, I wonder whether its
possible to save model orders to a seperate file
Thanks
nuncio
--
Nuncio.M
Research Scientist
National Center for Antarctic and Ocean research
Head land Sada
Vasco da Gamma
Goa-403804
[[alternative HTML version deleted]]
2011 Jun 28
2
A masked function is a masked function by any other name
Dear all,
It looks like I do not grasp the concept of masked functions enough as
to solve this trivial problem.
The code that replicates the problem (a source code tree that realizes
a R package actually) is under github so one can call it clone it
easily from the command line (though more experienced users will spot
the problem by browsing through the package code):
git clone http://jcborras at
2011 Apr 08
5
Avoiding a loop
Friends.
I cannot simplify this much, and I think the loop is unavoidable. As a
recovering C programmer I want to avoid loops and in cases like this I
almost allways can by using an apply function. But I suspect in this case
there is nothing I can do.
It is a finance example where a price series is compared to a moving
average. If the price goes above the average, plus a bit, buy the
2013 Mar 14
3
Working with string
Hello again,
Let say I have following string:
Vec <- c("sada", "asdsa", "sa")
Now I want to make each element of this vector with equal length.
Basically I want following vector:
c("sada ", "asdsa", "sa ")
Therefore we can get:
> nchar(c("sada ", "asdsa", "sa "))
[1] 5 5 5
Is there any
2011 Jun 08
1
The simplest bar graph with ggplot is difficult to realize
Dear all,
What is the simplest way of producing a bar graph using ggplot but
avoiding calling qplot?
That is, given:
d <- data.frame(x=seq(1,5), y=seq(1,5))
Why does the following line return an error?
ggplot(d, aes(x=x, y=y)) + stat_identity() + geom_bar(bindwidth=1)
Thanks in advance,
jcb!
2011 Jul 07
4
Return invisible list
Hi, I'm new to R. I'm trying to do some extreme value theory analysis,
looking at the Mean Excess Plot of a series. These are the commands I type
to get the plot:
x=read.table("data.txt",header=T)
goa=(x[,3])
meplot(goa)
I can see the plot, but I would like to see the values of the x and y axis.
According to this page
2011 Apr 20
2
user input
Dear users,
I have looked on different sources and found different functions to
prompt the user to provide input. However, I couldn't find one that does
exactly what I'm looking for.
select.list() and menu() are nice because a graphic window appears to
prompt the user. However, the user can only choose from a predefined
list of choices. readline() and scan() are more free in the
2011 Aug 19
4
Leading zeros
An embedded and charset-unspecified text was scrubbed...
Name: n?o dispon?vel
URL: <https://stat.ethz.ch/pipermail/r-help/attachments/20110819/e7764da5/attachment.pl>