Re: [R] Problems with Mann-Kendall trend test

[Morteza Firouzi via R-help]

Dear,
You have to store your data as a Time-Series (ts), first. To define a column of 
data as ts, you may use this:
library(timeSeries)Nile <- read.csv(file.choose(), header=F)#If your data is 
monthly, you may define the frequency as 12, for annual ts set freq. as 1.
#If your data starts from for e.g., 1990, then:NileTS <- ts(Nile, frequency=1, 
start=c(1990,1))#To plot the time seriesplot.ts(NileTS)
Morteza
 
 

On Saturday, June 18, 2016 3:47 AM, lily li  wrote:
 

 Dear R users,

Can anyone help me with mann-kendall trend test? I tried to use the newest
packages 'trend', and the function mk.test, but had problems in applying
the input data.

For example, res <- mk.test(Nile), Nile is a time-series data. But when I
use my dataset, with one column which is a time-series data, it says
"error: input must be ts object". How to do this? Thanks for your help.

    [[alternative HTML version deleted]]

__
R-help@r-project.org mailing list -- To UNSUBSCRIBE and more, see
https://stat.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide http://www.R-project.org/posting-guide.html
and provide commented, minimal, self-contained, reproducible code.


  
[[alternative HTML version deleted]]

__
R-help@r-project.org mailing list -- To UNSUBSCRIBE and more, see
https://stat.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide http://www.R-project.org/posting-guide.html
and provide commented, minimal, self-contained, reproducible code.

Re: [R] Help me

[Morteza Firouzi via R-help]

 Moses,
If I understand correctly, you are installed R and Rstudio.Please do find the 
package(s) you would like to use.Once you run rstudio, you can search and 
install the package(s) using the bottom right corner menu:Packages> Install> 
install from Repository (Cran)> search the package you would like to use in the 
box below 'Packages',then select the package name, and do not forget to let the 
default settings for "install dependencies", since you're new the R, and 
finally Install the package.
Once the package is downloaded and installed, you can see the name of the 
installed package in the same window. To run it, you have to check the box to 
load the package.You may find the documentation for each package using 
'help()', help(type the package name).For the rest, instead of using "Help me" 
in the subject line, please mention the package name and your problem, and 
explain exactly what would you do in the body.
Good Luck.Morteza
 

On Thursday, March 3, 2016 2:12 PM, Moss Moss  wrote:
 

 Please, I need help from you. I am new to R.

I installed R, Rstudio and R Crans or packages that I will like to
use. Thanks for your help on that.

Right now, what do I do? Will I upload the CRAN packages already
installed, so that a GUI interface will come up.

I am working on Location Model and need some help.

Moses

__
R-help@r-project.org mailing list -- To UNSUBSCRIBE and more, see
https://stat.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide http://www.R-project.org/posting-guide.html
and provide commented, minimal, self-contained, reproducible code.


  
[[alternative HTML version deleted]]

__
R-help@r-project.org mailing list -- To UNSUBSCRIBE and more, see
https://stat.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide http://www.R-project.org/posting-guide.html
and provide commented, minimal, self-contained, reproducible code.

[R] zyp Vs. kendall package for Time Series Trend

[Morteza Firouzi via R-help]

Dear members,
I need to detect trends in time series. To remove the effect of "Lag-1 serial 
correlation", it is suggested to use either Yue or Zhang method. Both 
methods are available in "zyp" package. The package uses "kendall" package for 
trend analysis.  


Based on Yue (2002), if the lag-1 serial correlation is significant, TFPW 
method will remove the effects of it prior to the trend test; otherwise trend 
test will be applied on original time series. 

I've compared the results of a sample time series with non-significant lag-1 
serial correlation, using both zyp & kendall packages. "yuepilon" method in 
"zyp" gives me the following results:tau: 0.075 & sig: 0.388
while "kendall" package gives me this: 
tau: 0.109 & sig: 0.216



The question is : Does "zyp" change the significance of the trend in this case 
as well? Is this a malfunction or did I miss something?
I've checked the script and it is mentioned (ln 65) : # Prewhiten the original 
series
c <- acf(data,lag.max=1,plot=FALSE,na.action=na.pass)$acf[2]


Thank you for your consideration.
Best regards,
Morteza





   __
R-help@r-project.org mailing list -- To UNSUBSCRIBE and more, see
https://stat.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide http://www.R-project.org/posting-guide.html
and provide commented, minimal, self-contained, reproducible code.

[R] missForest: Looping through files in a folder

[Morteza Firouzi via R-help]

Dear members,
Could you please help me on this issue. I've already searched and I watched 
some videos, but it was not useful.I need help to loop through the files in a 
folder (200+ csv files). I am using missForest() to impute missing values. If I 
run the code for each single file, I have to do as following:
## main script for each single file
G1334108 <- read.csv(file.choose(), header = T)

G1334108.F <- missForest(G1334108, verbose = TRUE, maxiter = 5)

write.csv(G1334108.F$ximp, file = 'G1334108_F.csv')

I tried these below script codes to loop the function before writing here:
# 1st tryall.files <- list.files()
   my.files <- grep(".*csv", all_files, value=T)
for(i in my.files){
# do your operations here
G1344108.Forest <- missForest(G1344108, verbose = TRUE, maxiter = 5)
# save
output.filename <- gsub("(.*?).csv", "\\1.csv", i)
write.table(G1344108.Forest$ximp, output.filename)
}##  2nd try
files <- list.files()lapply(files, function(x) {my.files <- read.csv("*.csv", 
header = T)missforest.out <- missForest(my.files, verbose = TRUE, maxiter = 
5)write.csv(missforest.out$ximp, file = '*_F.csv')
}Thank you for the time.Best regards,Morteza

[[alternative HTML version deleted]]

__
R-help@r-project.org mailing list -- To UNSUBSCRIBE and more, see
https://stat.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide http://www.R-project.org/posting-guide.html
and provide commented, minimal, self-contained, reproducible code.

Re: [R] missForest: Looping through files in a folder

[Morteza Firouzi via R-help]

Hi Jim,Thank you very much for the time.You saved me 3 days!
Best regards,
Morteza 

On Saturday, January 2, 2016 5:33 AM, Jim Lemon <drjimle...@gmail.com> 
wrote:
 

 Hi Morteza,What you may want is this:
my.files<-list.files(pattern=".csv")newfiles<-gsub(".","_F.",my.files,fixed=TRUE)for(i
 in 1:length(my.files)) { mydat<-read.csv(my.files[i]) 
mydatimp<-missForest(mydat,verbose=TRUE,maxiter=5) 
write.csv(mydatimp$ximp,newfiles[i])}
Jim

On Sat, Jan 2, 2016 at 5:32 AM, Morteza Firouzi via R-help 
<r-help@r-project.org> wrote:

Dear members,
Could you please help me on this issue. I've already searched and I watched 
some videos, but it was not useful.I need help to loop through the files in a 
folder (200+ csv files). I am using missForest() to impute missing values. If I 
run the code for each single file, I have to do as following:
## main script for each single file
G1334108 <- read.csv(file.choose(), header = T)

G1334108.F <- missForest(G1334108, verbose = TRUE, maxiter = 5)

write.csv(G1334108.F$ximp, file = 'G1334108_F.csv')

I tried these below script codes to loop the function before writing here:
# 1st tryall.files <- list.files()
   my.files <- grep(".*csv", all_files, value=T)
            for(i in my.files){
    # do your operations here
    G1344108.Forest <- missForest(G1344108, verbose = TRUE, maxiter = 5)
    # save
    output.filename <- gsub("(.*?).csv", "\\1.csv", i)
    write.table(G1344108.Forest$ximp, output.filename)
}##  2nd try
files <- list.files()lapply(files, function(x) {my.files <- read.csv("*.csv", 
header = T)missforest.out <- missForest(my.files, verbose = TRUE, maxiter = 
5)write.csv(missforest.out$ximp, file = '*_F.csv')
}Thank you for the time.Best regards,Morteza

        [[alternative HTML version deleted]]

__
R-help@r-project.org mailing list -- To UNSUBSCRIBE and more, see
https://stat.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide http://www.R-project.org/posting-guide.html
and provide commented, minimal, self-contained, reproducible code.



  
[[alternative HTML version deleted]]

__
R-help@r-project.org mailing list -- To UNSUBSCRIBE and more, see
https://stat.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide http://www.R-project.org/posting-guide.html
and provide commented, minimal, self-contained, reproducible code.