You should explain the context of your data/variable with greater detail. If your variable is the number of trades placed on a specific stock for a given day then they very well could be real zeros. However, if you're unsure if these are real zeros or if the zeros sometimes represent real zeros and other times represent missing observations then follow Nick's advice. Either way, the better of a description of your data and model that you provide the better your chances are of getting better advice.
HTH,
Billy
Sent from my iPhone
On Nov 21, 2012, at 6:33, Nick Cox <njcoxstata@gmail.com> wrote:
> If I understand you correctly, these zeros are not really zeros, but
> codes for missing values. I've never used -mgarch- at all, but I would
> try recoding to missing.
>
> See -help mvdecode-.
>
> Nick
>
> On Wed, Nov 21, 2012 at 2:28 PM, Hiroshi kameyama
> <kameyama.hiroshi@gmail.com> wrote:
>
>> I am not new for Stata, but new for MGARCH.
>>
>> I would like to apply MGARCH.
>> The dataset is as follows, daily data.
>> But not all days have actual value, "0" means no trade for this day.
>> Can I proceed to apply MGARCH?
>>
>> Your suggestion would be greatly appreciated.
>> Thank you in advance.
>>
>> Hiroshi KAMEYAMA
>> KAGAWA Univ., Japan
>>
>>
>> t soymeal soybean corn time
>> 1 0 6.7998916 5.857933 02jul2007
>> 2 0 6.7824755 5.823046 03jul2007
>> 3 5.462984 0 0 04jul2007
>> 4 5.4493202 6.7872811 5.836272 05jul2007
>> 6 5.4506092 0 0 07jul2007
>> 7 5.4548937 0 0 08jul2007
>> 8 0 6.8046145 5.858647 09jul2007
>> 9 0 6.824917 5.877736 10jul2007
>>
>> to
>> 1895
> *
> * 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/
