[R] Fw: Granger casuality test in r

Eneida Permeti permeti@eneid@ @ending from y@hoo@com
Fri Nov 30 18:37:04 CET 2018


 

   ----- Forwarded Message ----- From: Eneida Permeti <permeti.eneida using yahoo.com>To: John C Frain <frainj using gmail.com>Sent: Friday, November 30, 2018, 9:24:01 AM PSTSubject: Re: [R] Granger casuality test in r
  Dear JohnThank you for responding me.I have attached my data. I am studying the relationship between Public debt and economic growth.The time series of public debt is not stationary an I have differenced it.Than I have estimated a VAR model.But by the results of Granger causality test, I am afraid that something is wrong.Please can you help me?Best regardsEneida Permeti
    On Friday, November 30, 2018, 8:17:09 AM PST, John C Frain <frainj using gmail.com> wrote:  
 
 
On Fri, 30 Nov 2018 at 14:40, Eneida Permeti via R-help <r-help using r-project.org> wrote:


The results of my Granger causality test in r are below. VARp is my VAR model and I have two endogenous variables. From the results, I have only instantaneous causality. What does it mean?Thank you so much
> causality(VARp,cause="The.economic.growth")
$Granger

    Granger causality H0: The.economic.growth do not Granger-cause
    The.differenced.public.debt

data:  VAR object VARp
F-Test = 0.4038, df1 = 6, df2 = 8, p-value = 0.8573


$Instant

    H0: No instantaneous causality between: The.economic.growth and
    The.differenced.public.debt

data:  VAR object VARp
Chi-squared = 6.0964, df = 1, p-value = 0.01355


> causality(VARp,cause="The.differenced.public.debt")
$Granger

    Granger causality H0: The.differenced.public.debt do not Granger-cause
    The.economic.growth

data:  VAR object VARp
F-Test = 0.70214, df1 = 6, df2 = 8, p-value = 0.6572


$Instant

    H0: No instantaneous causality between: The.differenced.public.debt and
    The.economic.growth

data:  VAR object VARp
Chi-squared = 6.0964, df = 1, p-value = 0.01355
Inviato da Yahoo Mail su Android
        [[alternative HTML version deleted]]

______________________________________________
R-help using 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.


It appears that you have not found Granger causality.  I would not be surprised at this result.
You growth rate is almost equivalent to the log difference of GPD at constant prices. (real GDP). I suspect that your   The.differenced.public.debt is at current prices and is not log transformed.
Granger Causality requires you to control for other variables.  For example other variables may be causing both of your variables. If such is the case your finding of Granger Causality may be spurious.
3 Aranleigh Park
Rathfarnham
Dublin 14
Ireland
www.tcd.ie/Economics/staff/frainj/home.html
mailto:frainj using tcd.ie
mailto:frainj using gmail.com     


More information about the R-help mailing list