lagged variables

classic Classic list List threaded Threaded
4 messages Options
Reply | Threaded
Open this post in threaded view
|

lagged variables

saraberta
hi guys,
i have some trouble in creating lagged variables to use as external regressors.
i'm trying to use lag(x) but it gives me as result the same time series (x), adding this part at the end:

attr(,"tsp")
[1]    0 2323    1

where do i wrong?are there other functions to be used?
thanks
sara
Reply | Threaded
Open this post in threaded view
|

Re: lagged variables

Rui Barradas
Hello,

You are doing nothing wrong. Follow this example.

x <- ts(1:5)
# Seems the same with different start and end
lag(x)
# But it's not
cbind(x, lag(x))

Are there other functions? I know of at least one. (In the end.)
These two, though not lags, might also be of interess to you.

?window
y <- window(x, start=2, end=5)
cbind(x, y)

?window.zoo # package zoo

#
# Package: TSA
# Title: Time Series Analysis
# Version: 0.98
# Date: 2010-7-31
# Author: Kung-Sik Chan
`zlag` <-
function (x, d = 1)
{
     if (d != as.integer(d) || d < 0)
         stop("d must be a non-negative integer")
     if (d == 0)
         return(x)
     else return(c(rep(NA, d), rev(rev(x)[-(1:d)])))
}

cbind(x, zlag(x))

Hope this helps,

Rui Barradas

Em 25-07-2012 09:10, saraberta escreveu:

> hi guys,
> i have some trouble in creating lagged variables to use as external
> regressors.
> i'm trying to use lag(x) but it gives me as result the same time series (x),
> adding this part at the end:
>
> attr(,"tsp")
> [1]    0 2323    1
>
> where do i wrong?are there other functions to be used?
> thanks
> sara
>
>
>
>
> --
> View this message in context: http://r.789695.n4.nabble.com/lagged-variables-tp4637734.html
> Sent from the R help mailing list archive at Nabble.com.
>
> ______________________________________________
> [hidden email] mailing list
> 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.

______________________________________________
[hidden email] mailing list
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.
Reply | Threaded
Open this post in threaded view
|

Re: lagged variables

Gabor Grothendieck
In reply to this post by saraberta
On Wed, Jul 25, 2012 at 4:10 AM, saraberta <[hidden email]> wrote:

> hi guys,
> i have some trouble in creating lagged variables to use as external
> regressors.
> i'm trying to use lag(x) but it gives me as result the same time series (x),
> adding this part at the end:
>
> attr(,"tsp")
> [1]    0 2323    1
>
> where do i wrong?are there other functions to be used?
> thanks
> sara
>

The dyn and dynlm packages address this problem.

--
Statistics & Software Consulting
GKX Group, GKX Associates Inc.
tel: 1-877-GKX-GROUP
email: ggrothendieck at gmail.com

______________________________________________
[hidden email] mailing list
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.
Reply | Threaded
Open this post in threaded view
|

Re: lagged variables

Federman, Douglas
In reply to this post by saraberta
There is a Lag function in Hmisc and I found this on StackExchange

shift <- function (x, shift_by) { #similar to lag function
    stopifnot(is.numeric(shift_by))
    stopifnot(is.numeric(x))
    if (length(shift_by)>1)
        return(sapply(shift_by,shift, x=x))
    out<-NULL
    abs_shift_by=abs(shift_by)
    if (shift_by > 0 )
        out<-c(tail(x,-abs_shift_by),rep(NA,abs_shift_by))
    else if (shift_by < 0 )
        out<-c(rep(NA,abs_shift_by), head(x,-abs_shift_by))
    else
        out<-x
    out
}

-----Original Message-----
From: [hidden email] [mailto:[hidden email]]
On Behalf Of saraberta
Sent: Wednesday, July 25, 2012 4:10 AM
To: [hidden email]
Subject: [R] lagged variables

hi guys,
i have some trouble in creating lagged variables to use as external
regressors.
i'm trying to use lag(x) but it gives me as result the same time series
(x), adding this part at the end:

attr(,"tsp")
[1]    0 2323    1

where do i wrong?are there other functions to be used?
thanks
sara




--
View this message in context:
http://r.789695.n4.nabble.com/lagged-variables-tp4637734.html
Sent from the R help mailing list archive at Nabble.com.

______________________________________________
[hidden email] mailing list
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.

______________________________________________
[hidden email] mailing list
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.