Adam,
You need equally spaced data for ordinary time series, if
your acf and pacf are to provide correct significance tests.
Moreover, the ordinary ADF test will require such data as well.
If you have unequally spaced data, you should ask yourself
whether you can responsibly interpolate. If you can, you will
need to do so. If you cannot, then you will have to use another
method that permits unequally spaced data.
Regards,
Robert Yaffee
Robert A. Yaffee, Ph.D.
Research Professor
Shirley M. Ehrenkranz
School of Social Work
New York University
home address:
Apt 19-W
2100 Linwood Ave.
Fort Lee, NJ
07024-3171
Phone: 201-242-3824
Fax: 201-242-3825
[email protected]
----- Original Message -----
From: adam dvir <[email protected]>
Date: Sunday, September 17, 2006 3:32 pm
Subject: RE: st: RE: ADF test for seasonaly data
> Hi Jason,
>
> Unfortunaly it doesn't. I thought maybe I can do seasonally
> differences for
> the change relative to last season (e.g. April 2001 - April 2000
> etc.). But
> I find Auto-Correlation problem that does not seems to fade using
> lagged-differences.
>
> what do you think?
> thanks,
>
> Adam
>
>
> >From: "Jason Yackee" <[email protected]>
> >Reply-To: [email protected]
> >To: <[email protected]>
> >Subject: st: RE: ADF test for seasonaly data
> >Date: Sun, 17 Sep 2006 11:50:36 -0700
> >
> >Does it make sense to make your own 7-month "year" so that the
> lag of
> >October is the previous April, &c?
> >
> >Jason
> >
> >-----Original Message-----
> >From: [email protected]
> >[mailto:[email protected]] On Behalf Of adam dvir
> >Sent: Sunday, September 17, 2006 11:46 AM
> >To: [email protected]
> >Subject: st: ADF test for seasonaly data
> >
> >Hello everyone,
> >
> >I have a seasonally data of prices and quantities of Avocado
> export to
> >Europe, that I want to check for stationarity using DF or ADF tests.
> >
> >the problem is that my data is not continuous, and includes only the
> >monthes
> >october to April (7 monthes), when the other monthes are blank
> (missing>values). When I use the ADF regression (with only 1
> lagged diference), I
> >
> >automatically lose 36 obs. of my 123 obs., because of the data's
> missing>
> >monthes (I have obs. for 18 years). Moreover, each lag I add to the
> >rerassion makes one month to be ommited from the regrssion (for
> example:>1
> >lag - October is omitted, 2 lags - November is omitted etc.)
> >
> >Is it right to use the ADF test? Or is there other test for
> stationarity>
> >that might feet better to my case?
> >
> >
> >
> >Thank you very much,
> >
> >
> >
> >Adam Dvir
> >
> >_________________________________________________________________
> >Check the weather nationwide with MSN Search: Try it now!
> >http://search.msn.com/results.aspx?q=weather&FORM=WLMTAG
> >
> >*
> >* For searches and help try:
> >* http://www.stata.com/support/faqs/res/findit.html
> >* http://www.stata.com/support/statalist/faq
> >* http://www.ats.ucla.edu/stat/stata/
> >
> >*
> >* For searches and help try:
> >* http://www.stata.com/support/faqs/res/findit.html
> >* http://www.stata.com/support/statalist/faq
> >* http://www.ats.ucla.edu/stat/stata/
>
> _________________________________________________________________
> Check the weather nationwide with MSN Search: Try it now!
> http://search.msn.com/results.aspx?q=weather&FORM=WLMTAG
>
> *
> * For searches and help try:
> * http://www.stata.com/support/faqs/res/findit.html
> * http://www.stata.com/support/statalist/faq
> * http://www.ats.ucla.edu/stat/stata/
>
*
* For searches and help try:
* http://www.stata.com/support/faqs/res/findit.html
* http://www.stata.com/support/statalist/faq
* http://www.ats.ucla.edu/stat/stata/
