Bookmark and Share

Notice: On April 23, 2014, Statalist moved from an email list to a forum, based at statalist.org.


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

Re: st: count time series model


From   Robert A Yaffee <[email protected]>
To   [email protected]
Subject   Re: st: count time series model
Date   Wed, 29 Jan 2014 13:11:48 -0500

Ikuho,
   You're quite welcome.  Have you thought of trying  a modification
of xtpoisson ?
          Bob

On Wed, Jan 29, 2014 at 12:50 PM, Ikuho  Kochi <[email protected]> wrote:
> Thank you Bob for your suggestion.  I looked over the paper and it seems more complicated than I hoped for.  Not sure if I am able to
> carry out this estimation...
>
> I still wonder if there is any ready to use built-in command in STATA to test and account for serial correlation in count time series data?
>
> ikuho
> ________________________________________
> From: [email protected] [[email protected]] on behalf of Robert A Yaffee [[email protected]]
> Sent: Tuesday, January 28, 2014 9:50 PM
> To: [email protected]
> Subject: Re: st: count time series model
>
> Ikuho,
>      You may want to investigate the Generalized Autoregressive Score models of
> Creele, Koopman and Lucas, which implement this kind of analysis.
>       Sincerely,
>            Bob Yaffee
>
> On Tue, Jan 28, 2014 at 4:42 PM, Ikuho  Kochi <[email protected]> wrote:
>>
>> I am using STATA version 12, not 13. Sorry for the confusion.
>>
>> On Jan 28, 2014, at 10:39 AM, "Ikuho  Kochi" <[email protected]> wrote:
>>
>>> Dear fellow statisticians,
>>>
>>> I have a count time series data set and would like to test and correct serial correlations in my time series model.
>>> I saw previous posts related to this topic, and one of the comment was to use Newey-West standard errors with nbreg command.
>>> However, I can not find the option to incorperate Newey-West standard errors with nbreg command with STATA version 13.
>>> I wonder if somebody knows how to do this, or have any other suggestion.
>>>
>>> Also, I wonder if there is any test that is already programmed in STATA to test for serial correlation in count time series data model.
>>> I have more than 500 subset of samples, and it would be very time consuming for me to check the correlation between residuals for different time period for each of 500 samples...
>>> If there is already a programmed command for this test, it will help me to save time greatly.
>>>
>>> Thank you for your attention.
>>>
>>> Ikuho
>>>
>>>
>>> *
>>> *   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/
>
>
>
> --
> Bob Yaffee
>
> Biosketch: http://homepages.nyu.edu/~ray1/Biosketch2009.pdf
>
> CV:  http://homepages.nyu.edu/~ray1/vita.pdf
> *
> *   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/



-- 
Bob Yaffee

Biosketch: http://homepages.nyu.edu/~ray1/Biosketch2009.pdf

CV:  http://homepages.nyu.edu/~ray1/vita.pdf
*
*   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 1996–2018 StataCorp LLC   |   Terms of use   |   Privacy   |   Contact us   |   Site index