Statalist


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

st: Re: temporal disaggregation - NOW -denton-


From   Kit Baum <baum@bc.edu>
To   statalist@hsphsun2.harvard.edu
Subject   st: Re: temporal disaggregation - NOW -denton-
Date   Thu, 30 Apr 2009 06:40:20 -0400

<>
It does not matter whether you use logs or not. -denton- forces the constructed quarterly estimates to add up to the annual estimate. Whether you express the annual estimate in log or level terms does not affect that algebra.

You can only use one indicator, but if you had more than one, you could combine them with principal components. If you had three, form the principal components of those three variables and use the first one as the indicator.


Kit Baum   |   Boston College Economics & DIW Berlin   |   http://ideas.repec.org/e/pba1.html
An Introduction to Stata Programming | http://www.stata-press.com/books/isp.html
   An Introduction to Modern Econometrics Using Stata  |   http://www.stata-press.com/books/imeus.html



On Apr 30, 2009, at 02:33 , statalist-digest wrote:

I found this old post on the web that seems to help me on the question I sent yesterday - I need to interpolate GDP data from a quarterly to a monthly frequency.

I have downloaded the -denton- command and used it. I am however not sure my results are correct. I am using an index of industrial production as an indicator.

Shall I express GDP and industrial production in logs or not when using denton? Or are the results to denton independent of whether the variables are in logs or not?

Also, it appears that only one indicator can be used. Is there a way to use two different indicators?

*
*   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