Bookmark and Share

Notice: On March 31, it was announced that Statalist is moving from an email list to a forum. The old list will shut down on April 23, and its replacement, statalist.org is already up and running.


[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: st: VECM: Deterministic Factors & Significance for ce1 in short-run parameters


From   Prakash Singh <prakashbhu@gmail.com>
To   statalist@hsphsun2.harvard.edu
Subject   Re: st: VECM: Deterministic Factors & Significance for ce1 in short-run parameters
Date   Fri, 8 Feb 2013 00:25:35 +0530

Daniel
You are not wrong but if you read the literature little carefully
significant negative coefficient of ce is more authentic proof of
presence of cointegration.
error correction can only happen in presence of cointgration, going by
this argument significant negative coefficient of ce proves  presence
of cointegration. the size of ce coefficient*100 is percent of
correction happening in the next period (year, quarter or so) and the
percent of correction (coming to long run equilibrium position from
deviation) will for sure be in 0 to 100 bound and thus coefficient
will be between 0 to 1 and negative sign represents return towards
long run. Well, I would not suggest you to go for
2-Step-Eagle-Granger-Approach in multivariate case.

Prakash

On Thu, Feb 7, 2013 at 9:45 PM, Daniel Ilg <daniel.ilg@ku.de> wrote:
> Dear Prakash,
>
> Thank you for your answer! I thought, a cointegration relation proven through Johannsen or 2-Step-Eagle-Granger-Approach (or other tests of the residuals) is sufficient. Am I wrong? I read as well that the cointegrating coefficient should be negative and between 0 and 1. Do you have a clue why?
>
> Best regards,
> Daniel
>
> Mit freundlichen Grüßen/Best regards
>
> Daniel Ilg
> wissenschaftlicher Mitarbeiter/Research Associate
> Lehrstuhl für Wirtschaftstheorie/Chair of Economic Theory
>
> Katholische Universität Eichstätt-Ingolstadt/Catholic University Eichstaett-Ingolstadt
> Auf der Schanz 49
> 85049 Ingolstadt
> Tel.: +49 841 937 1857
> Mobil: +49 171 653 113 1
> E-Mail: daniel.ilg@ku.de
>
>
> -----Original Message-----
> From: owner-statalist@hsphsun2.harvard.edu [mailto:owner-statalist@hsphsun2.harvard.edu] On Behalf Of Prakash Singh
> Sent: Thursday, February 07, 2013 4:58 PM
> To: statalist@hsphsun2.harvard.edu
> Subject: Re: st: VECM: Deterministic Factors & Significance for ce1 in short-run parameters
>
> Dear Daniel
>
> As far as my knowledge goes significant ce term is essential for cointegration; in other words significant ce terms indicate presence of cointegration and the size of cointegrating coefficient is supposed to be negative and should be in between 0 to 1.
>
>
> Hope it helps
>
> Prakash
> On Wed, Feb 6, 2013 at 7:04 PM, Daniel Ilg <daniel.ilg@ku.de> wrote:
>> Dear users,
>>
>> I have 1 problem and 1 question.
>> I apply the VECM for a time series. My dependent variable is default
>> probability and all variables are I(1).
>>
>> My problem:
>> I want to determine the deterministic components for my VECM.
>> My independent variables have different deterministic components, so I
>> can't derive my deterministic factors from theory. Another problem is,
>> that I'm just interested in a cointegration relation of 1 (default
>> probability is the variable I want to explain). So I'm not interested
>> in more ranks than 1. I applied pentula and modified pentula principle
>> but the problem is, I get more ranks and - as said before - I'm just
>> interested in 1 cointegration relation. So, how can I determine the
>> deterministic components of my VECM for rank = 1?
>>
>> My question:
>> When I get my VEC output does my "ce1" need to be significant in my
>> Error Correction "d_default_probability"? What does it mean when "ce1"
>> is not significant in the short-run?
>>
>> Thx a lot & regards,
>> Daniel
>>
>>
>> *
>> *   For searches and help try:
>> *   http://www.stata.com/help.cgi?search
>> *   http://www.stata.com/support/faqs/resources/statalist-faq/
>> *   http://www.ats.ucla.edu/stat/stata/
> *
> *   For searches and help try:
> *   http://www.stata.com/help.cgi?search
> *   http://www.stata.com/support/faqs/resources/statalist-faq/
> *   http://www.ats.ucla.edu/stat/stata/
>
>
> *
> *   For searches and help try:
> *   http://www.stata.com/help.cgi?search
> *   http://www.stata.com/support/faqs/resources/statalist-faq/
> *   http://www.ats.ucla.edu/stat/stata/

*
*   For searches and help try:
*   http://www.stata.com/help.cgi?search
*   http://www.stata.com/support/faqs/resources/statalist-faq/
*   http://www.ats.ucla.edu/stat/stata/


© Copyright 1996–2014 StataCorp LP   |   Terms of use   |   Privacy   |   Contact us   |   Site index