st: Réf.: Re: st: bootstrap a test linear hypotheses

[Thierry Hounsa <hounsathierry@yahoo.fr>]

Thanks for your reply. What are the ideas in the litterature on how to do it?



------------------------------
Le mar. 4 juin 2013 16:32 HAP, Nick Cox a écrit :

>The problem is in the idea. How do you bootstrap a time series? Once
>you have sampled a set of observations with replacement, how do you
>then define the time order for such a sample, with typically some
>observations repeated and some missing? There are ideas in the
>literature on how to do it, but -bootstrap- does not implement them.
>Nick
>njcoxstata@gmail.com
>
>
>On 5 June 2013 00:23, Thierry Hounsa <hounsathierry@yahoo.fr> wrote:
>> Hi all.. How to bootstrap a test linear hypothese after the estimation of a Vectorial Autoregressive Model?
>> Indeed, I perform these commands:
>>
>> *** estimate a VAR of 2 lags on the variables Y1 and Y2***
>> var Y1 Y2
>> *** Perform a test on the coefficients of Y1***
>> test ([Y1]: L.Y2 L2.Y2)
>> *** bootstrap the previous test***
>> bootstrap, reps(50) : test ([Y1]: L.Y2 L2.Y2)
>> But I get the following error message: "time-series operators are not allowed with bootstrap without panels, see tsset, r(198)"
>> What's wrong with my commands? What should I do?
>> Thanks in advance
>>
>> *
>> *   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/