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: ETA Stata 12.0


From   DE SOUZA Eric <[email protected]>
To   "[email protected]" <[email protected]>
Subject   RE: st: ETA Stata 12.0
Date   Thu, 7 Apr 2011 17:03:48 +0200

The vecrank command in Stata 11.1 allows for seasonal indicators with the sindicators option.
I used to use RATS several years ago, then switched to PcGive (OxMetrics) for dynamic linear systems modelling. Easier to teach from, too.
My students have used JMulti for SVARs in their Master's theses.


Eric de Souza

-----Original Message-----
From: [email protected] [mailto:[email protected]] On Behalf Of Charles Koss
Sent: 07 April 2011 16:22
To: [email protected]
Subject: Re: st: ETA Stata 12.0

Good to know. In addition to that, VECRANK command does not allow the inclussion of seasonal terms as well as structural breaks either in levels or trend.

The following is worth reading about what is needed in Stata:
Yafee, R.A  2007. Stata 10 (Time Series and Forecasting). Journal of Statistical Software. 23(1):1-18.
http://www.jstatsoft.org/v23/s01/paper

In conclusion, it seems to me that Stata is heading to a different market of researchers (stat. methods) than Eviews or Rats or open source such as Jmulti or GRETL.

Sincerely,

Charles

--
Charles Koss
http://charlesonnet.blogspot.com



On Thu, Apr 7, 2011 at 3:29 AM, DE SOUZA Eric
<[email protected]> wrote:
> In this connection, cointegration tests can be done with the introduction of stationary regressors (see Rahbek and Mosconi (1999), Cointegration rank inference with stationary regressors in VAR models, Econometrics Journal, which also refers to the relevant tables).  -vec- does not allow  the introduction of exogenous variables. It should be allowed with the mention that the critical values produced are not valid.
>
> Also, -var- does not omit a variable when its coefficient is constrained to zero. It produces a very tiny number instead (e-18 or e-19) with standard errors and all the rest.
>
> Reference is to Stata 11.1
>
>
> Eric de Souza
> College of Europe
> Brugge (Bruges), Belgium
> http://www.coleurope.eu
>
>
> -----Original Message-----
> From: [email protected] [mailto:[email protected]] On Behalf Of Charles Koss
> Sent: 06 April 2011 21:17
> To: [email protected]
> Subject: Re: st: ETA Stata 12.0
>
> Well well! Since starting to use Stata, I am very impressed how much have learnt. Since we always want to purchase better products, I hope Stata 12 can produce complete inference for Vector Error Correction Models (vec command).
>
> Currently, Stata 10 and 11 can produce the estimates for the impulse response functions but it CAN NOT estimate the corresponding standard errors. So, those estimates are useless since WE DO NEED the standard errors for interpreting the results STATISTICALLY. I was amazed to learn that OPEN SOURCE software can do that job very efficiently. I would like to see this feature in STATA 12.
>
> With this feature, I can convince myself and the administrators that the purchase of Stata 12 is worthy, although version 11 is much better than 10. If someone has information that contradicts my opinion, please do share it.  See:
> http://www.stata.com/statalist/archive/2011-03/msg01435.html
>
> Sincerely,
>
> Charles
>
> --
> Charles Koss
> http://charlesonnet.blogspot.com
>
> On Wed, Apr 6, 2011 at 12:55 PM, Airey, David C <[email protected]> wrote:
>> .
>>
>> Personally, I would not mind getting another year out of Stata 11!
>>
>> -Dave
>>
>>
>>
>>> This was asked also on 17 March. See
>>>
>>> <
>>> http://www.stata.com/statalist/archive/2011-03/msg01113.html
>>> >
>>>
>>> The history of release dates is covered in
>>>
>>> <
>>> http://www.stata.com/support/faqs/res/history.html
>>> >
>>>
>>> You might like to build your model from the data....
>>>
>>> Here are two precedents from history:
>>>
>>> 1. Precisely when a release will be issued will be announced on this
>>> list (and also now on Facebook, Twitter, the Stata blog, etc.)
>>> perhaps a week before the actual date.
>>>
>>> 2. Precisely what will be included will also be announced then, and not before.
>>>
>>> These of course are just my personal generalisations, and StataCorp
>>> might do things differently for the next release.
>>>
>>> Otherwise speculation might be fun, but it is also likely to be
>>> completely futile! As the Tao Te Ching says, more or less, those who
>>> know do not say; those who say do not know.
>>>
>>> (I don't know either, so any and all requests for private information
>>> will also be futile.)
>>>
>>> Nick
>>
>> *
>> *   For searches and help try:
>> *   http://www.stata.com/help.cgi?search
>> *   http://www.stata.com/support/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/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/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/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/statalist/faq
*   http://www.ats.ucla.edu/stat/stata/


© Copyright 1996–2018 StataCorp LLC   |   Terms of use   |   Privacy   |   Contact us   |   Site index