Statalist


[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

st: RE: How can estimate dicount rate for each household in the panel data


From   "Nick Cox" <n.j.cox@durham.ac.uk>
To   <statalist@hsphsun2.harvard.edu>
Subject   st: RE: How can estimate dicount rate for each household in the panel data
Date   Tue, 17 Feb 2009 18:35:48 -0000

I wouldn't use -nl- to estimate this rate. On the face of it, it is more
natural to take logs and then fit a regression in each panel. You can do
that using -statsby- to pick up the regression slopes. Then -merge- back
to your original dataset. 

Nick 
n.j.cox@durham.ac.uk 

Quang Nguyen

I have the following sample data set:

 Household       x       y       t
 1                   60     120    7
 1                   120   100   30
 1                   100   300   60
 1                    50    250  900
 ...
 2
 2
 2
 ...
 I would like to estimate the discount rate for each household based
on the fomular: x = y*exp(-rt), where r is the interest rate. I know
that in Stata we can use: nl (x=y*(exp(*-{r}*t) for each household. My
questions are:

1. How we can ask State to do the the above estimate for all household
at the same time instead of one by one?

2. How can we save the ESTIMATED discount rates as an variable that
can be used for latter analysis? For instance, on emay study the
relationship between the household's discount rate and the educational
level.


*
*   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 1996–2014 StataCorp LP   |   Terms of use   |   Privacy   |   Contact us   |   What's new   |   Site index