Re: tsDyn package (TVAR usage)

207 views
Skip to first unread message

Matthieu Stigler

unread,
Jul 28, 2011, 7:44:32 AM7/28/11
to fra....@hotmail.it, ts...@googlegroups.com
Dear Francesca

Could you be more detailed on your example? Also, using artificial data?
It is otherwise difficult to understand what is exactly your problem.

The thVar argument is indeed designed to accept an external variable as
threshold variable, so it seems to be that what you are doing is
correct. Be sure however to include the name of the argument, to avoid
any conflict with other arguments, i.e. thVar=sspread instead of
sspread. Also, you do not need to specify the mTh argument in this case.

Hope this helps

Mat

PS: I am ccing to the package mailing list, to which you are welcome to
subscribe to.

Le 27/07/2011 11:01, fra....@hotmail.it a �crit :
> Hi,
> I'm using the tsDyn package to estimate a multivariate threshold VAR. Currently, I'm in trouble with the selection of an external transition variable. I cannot select it properly. My xls spreadsheet contains 3 variables: I want to use the first and the second one as "data" and the third one as a threshold (external). Among my attemps:
> 1) (Note: I relied on 2 different xls)
> rm(list=ls())
> library(tsDyn)
> usrates<- read.csv(file="usrates_r.csv", head=TRUE,sep=";")
> sspread<- read.csv(file="sspread.csv", head=TRUE,sep=";")
> tvar<- TVAR(usrates, lag=7, nthresh=2, thDelay=1, sspread, trim=0.1, mTh=3, plot=TRUE)
> class(tvar)
> summary(tvar)
>
> 2)
> tvar<- TVAR(usrates[,1:2], lag=7, nthresh=1, thDelay=1,usrates[,3],trim=0.1,plot=TRUE)
>
> Can you shed more lights on the "thVar" usage?
> Many thanks,
> Francesca

long...@gmail.com

unread,
Apr 7, 2017, 7:54:27 PM4/7/17
to tsdyn, fra....@hotmail.it
Dear Matthieu,
First I would like to thank you for your efforts in making and updating tsDyn. It's been very useful to me.
I've tried Francesca second specification and find it's working fine with a TVAR function. My concern is, there doesn't seem to be a similar option for the TVECM function, as they should basically have similar arguments for the transition variable.
What I mean is in the TVECM output, we don't explicitly know what variable does the threshold estimate represents, unlike in the TVAR which we can choose with option "thVar".

Regards,

Matthieu S

unread,
Apr 7, 2017, 7:57:42 PM4/7/17
to ts...@googlegroups.com, long...@gmail.com, fra....@hotmail.it
Hi

In TVECM, the threshold variable is the ECT itself, and there is no option to have an external transition variable. In the TVAR on the other side, the transition variable is either a (combination of ) lag(s), or an external variable.

Hope this helps,

Matthieu

--
You received this message because you are subscribed to the Google Groups "tsdyn" group.
To unsubscribe from this group and stop receiving emails from it, send an email to tsdyn+unsubscribe@googlegroups.com.
To post to this group, send email to ts...@googlegroups.com.
Visit this group at https://groups.google.com/group/tsdyn.
For more options, visit https://groups.google.com/d/optout.

Reply all
Reply to author
Forward
0 new messages