st: Re: Features for Stata 14

["Liu,Zhong" <zhong1969@hotmail.com>]

Dear Maarten:

Thanks and especially for reminding me about Stata. I should know it. Sorry.

Best,
Zhong

-----原始邮件----- 
From: Maarten Buis
Sent: Thursday, September 05, 2013 5:50 PM
To: statalist@hsphsun2.harvard.edu
Subject: Re: st: Re: Features for Stata 14

Ok, however expect numerical integration inside a simulation to be
painfully slow. Do you mean numerical integration _by_ simulation? If
so, take a look at this special issue of the Stata Journal:
http://www.stata-journal.com/sj6-2.html

Let me repeat that there is no such thing called STATA. Stata,
however, does exist. You may think this difference is trivial, but
people get upset about this on this list as it shows you have not
taken the time to read the Statalist FAQ and it upsets me because you
ignored a gentle hint.

-- Maarten


On Thu, Sep 5, 2013 at 11:24 AM, Liu,Zhong <zhong1969@hotmail.com> wrote:
> Dear Maarten:
>
> Thanks for your reply. But I do mean -normal(x)- rather 
> than -normalden(x)-.
> The issue involved is that I want to employ Equation (5.7) at p.126 of a
> paper authored by Kenneth A. Small and Harvey S. Rosen (1981) in
> Econometrica, Vol. 49, No. 1 (Jan., 1981), pp. 105-130. The paper is 
> titled
> "Applied Welfare Economics with Discrete Choice Models", which you can 
> find
> it by following the weblike below:
>
> http://www.jstor.org/discover/10.2307/1911129?uid=3737800&uid=2129&uid=2&uid=70&uid=4&sid=21102620779363
>
> And I also think that it is better for STATA to have an observation-level
> numerical integration since many simulation-based estimates need the
> functionality of the kind.
>
> Best,
> Zhong
>
> -----原始邮件----- From: Maarten Buis
> Sent: Thursday, September 05, 2013 3:48 PM
> To: statalist@hsphsun2.harvard.edu
> Subject: Re: st: Re: Features for Stata 14
>
> On Thu, Sep 5, 2013 at 8:40 AM, Liu,Zhong  wrote:
> > A dataset has two variables, say ‘v1’ and ‘v2’. Either of them has 
> > ‘N’
> > number of observations. I want to generate a third variable, say ‘v3’. 
> > The
> > 1st element of ‘v3’ is the numerical integration with the 1st element 
> > of
> > ‘v1’ as the lower limit to the integral and the 1st element of ‘v2’ 
> > as the
> > upper limit to the integral. Similarly, the 2nd element of ‘v3’ is the
> > numerical integration with the 2nd element of ‘v1’ as the lower limit 
> > to
> > the
> > integral and the 2nd element of ‘v2’ as the upper limit. And so on. The
> > function integrand is known STATA function such as 'normal(x)' or defined
> > by
> > a program written by users.
>
>
> For many distributions there is no need to do this numerically. In
> your example, I suppose you mean -normalden(x)- for the function that
> is to be integrated instead of -normal(x)-. -normalden()- gives you
> the PDF of the normal distribution (the well known bell shaped curve),
> while -normal()- gives you the CDF (an S-shaped curve). In that case
> the solution is straightforward, as you can see in the example below.
> In -help density_functions- you can find a list of distributions for
> which this solution has been implemented.
>
> *------------------ begin example ------------------
> // create some example data
> clear
> input v1 v2
> -2 2
> -1 1
> end
>
> // compute your integrals
> gen v3 = normal(v2) - normal(v1)
>
> // admire the result
> list
> *------------------- end example -------------------
> * (For more on examples I sent to the Statalist see:
> * http://www.maartenbuis.nl/example_faq )
>
> Hope this helps,
> Maarten
>
> Ps. As you can read in the Statalist FAQ
> (http://www.stata.com/support/faqs/resources/statalist-faq/#spell) it
> is Stata not STATA.
>
> ---------------------------------
> Maarten L. Buis
> WZB
> Reichpietschufer 50
> 10785 Berlin
> Germany
>
> http://www.maartenbuis.nl
> ---------------------------------
>
> *
> *   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/



--
---------------------------------
Maarten L. Buis
WZB
Reichpietschufer 50
10785 Berlin
Germany

http://www.maartenbuis.nl
---------------------------------

*
*   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/