The issue is discussed in Ronald Kessler and David Greenberg, LINEAR PANEL ANALYSIS: MODELS OF QUANTITATIVE CHANGE (Academic Press, 1981). In a nutshell, the two models are equivalent. Take the equation Y_t = a + b_1*Y_{t-1}+b_2*X_t + e_t. Subtract Y_{t-1} from both sides, which changes nothing. Now the dependent variable is a change score. The only difference is that the coefficient of Y_{t-1} is now b_1 - 1 instead of b_1. For certain types of analyis one version is more convenient than the other, but algebraicly they are interchangeable. David Greenberg, Sociology Department, New York University
----- Original Message -----
From: Clive Nicholas <[email protected]>
Date: Tuesday, February 24, 2004 7:48 pm
Subject: st: Lagged regression DVs: gross or net?
> All,
>
> When running regression models with lagged terms, is it better to
> use a
> dependent variable measured by _levels_ or by _first differences_?
> Or does
> it not matter?
>
> I ask since the use of any RHS lagged term transforms its
> interpretationinto one where a short-term process of adjustment is
> being described.
>
> Thanks.
>
> CLIVE NICHOLAS |t: 0(44)191 222 5969
> Politics Building |e: [email protected]
> School of Geography, |f: 0(44)870 126 2421
> Politics & Sociology |
> University of |
> Newcastle-upon-Tyne |
> Newcastle-upon-Tyne |
> NE1 7RU |
> United Kingdom |http://www.ncl.ac.uk/geps
> *
> * For searches and help try:
> * http://www.stata.com/support/faqs/res/findit.html
> * http://www.stata.com/support/statalist/faq
> * http://www.ats.ucla.edu/stat/stata/
>
*
* For searches and help try:
* http://www.stata.com/support/faqs/res/findit.html
* http://www.stata.com/support/statalist/faq
* http://www.ats.ucla.edu/stat/stata/
