Re: st: Fractional Response Model with Dynamic Component

[Nick Cox <njcoxstata@gmail.com>]

Minimal references that are name (date) are deprecated on Statalist,
as is explained in the Statalist FAQ that you were asked to read
before posting.

The main idea you refer to has much longer roots than your references
imply. That apart, I can't see any reason to regard a time component
as out of order. The point is just that bounded responses are
appropriately dealt with by ensuring that predictions are also bounded
by the use of an appropriate link function: not everyone uses that
terminology.

Jeff Wooldridge (many dates) is an occasional poster on this list. If
you're lucky, he may add a comment.

Nick

On Mon, Nov 12, 2012 at 9:50 PM, Anna-Leigh Stone
<alstone@crimson.ua.edu> wrote:
> Hello Statalist,
>
> I am interested in using the Papke and Wooldridge (2008) fractional
> response model. I have a dependent variable that lies within the zero
> and one range for the dependent variable with values that lie at zero
> but not one. However, the model will not converge without a dynamic
> component added in. I am using quarterly data so that there is a need
> for lagged values of the dependent variable. I know that Loudermilk
> (2007) added in a dynamic component into the Tobit model. Is there an
> appropriate way to add in a dynamic component into the fractional
> response model?
>
> Thank you for any help you can provide,
>
> Anna-Leigh
>
>
>
> --
> Anna-Leigh Stone
> Finance PHD Candidate 2014
> MA/EC 2011
> MSF 2010
> alstone@crimson.ua.edu
> *
> *   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/