Re: Gregory Hansen test [R-code]
Matthieu Stigler
I implemented recently the Gregory Hansne code for the package tsDyn,
but it is still in the development version, hence not in CRAN.
If you want to use it, I can send it to you but you will need to compile
the package yourself, with R CMD INSTALL, can you?
Matthieu
alessandr...@email.it a �crit :
> Dear. M. Stigler,
> some days ago I was seeking cointegration tests that deal with multiple breaks and I found your replies in the topic upon the "Cointegration with multiple structural breaks". They have been really helpful for me, thank you!
> Now I am trying to apply the Gregory-Hansen test to my series in R, I checked online without finding anything more than a Gauss code. I am wondering if you can help or suggest me a link where I can find an R code.
>
> Thank you,
>
> best regards,
>
> Alessandro Carraro
>
Matthieu Stigler
Please find the just updated source code at:
http://github.com/MatthieuStigler/tsDyn/tree/gregHanTest
and use the "download source". To compile, you will find many
informations on the net, on windows you will need to install Rtools.
Please be aware that this is still an untested code. It might contain
errors. There is an experimental "hpc" argument, whihc allows to run in
parallel. If you have some time, you are welcome to test it (see
drelated documentation on foreach) and give feedback.
May I ask on what of application you are applying this test?
Matthieu
PS: please answer too on the tsdyn mailing list.
Alessandro Carraro a �crit :
> Matthieu,
> thank you so much for answering, I've never done this procedure but I
> think I'll be able.
> Thanks again for you help, I'm looking forward to hearing from you.
> Best regards,
>
> Alessandro
>
>
>
>
> ----- Original Message ----- From: "Matthieu Stigler"
> <matthieu...@gmail.com>
> To: <alessandr...@email.it>; <ts...@googlegroups.com>
> Sent: Thursday, May 27, 2010 9:03 AM
> Subject: Re: Gregory Hansen test [R-code]
Matthieu Stigler
Note that the package tsDyn contains a number of features to study
price transmission, notably the threshold cointegration model used in
numerous price transmisison studies. It might be useful for you!
May I ask where you are working?
Matthieu
Alessandro Carraro a �crit :
> Dear Matthieu,
> thanks for linking me the source code, I will follow your indications
> for the installation.
> Actually I'm studying price transmission among some food chains, but
> during my analysis I encountered some problems (related to a correct
> computation of the cointegrating relationships) generated by the
> presence of some structural breaks, hence to look more in deep the
> cointegration relationships between the time series I have decided to
> apply tests that take in consideration even the structural changes.
> Thank you again for your support and for helping all the R-community
> with tsDyn package, if I test the "hpc" of course I will send you a
> feedback.
>
> Best regards,
>
> Alessandro
>
>
> ----- Original Message ----- From: "Matthieu Stigler"
> <matthieu...@gmail.com>
> To: "Alessandro Carraro" <alessandr...@email.it>;
> <ts...@googlegroups.com>
> Sent: Friday, May 28, 2010 4:12 PM
hgar...@ubiobio.cl
Could you send me the program to use the contrast? Please.
Regards,
Héctor.