Bookmark and Share

Notice: On March 31, it was announced that Statalist is moving from an email list to a forum. The old list will shut down on April 23, and its replacement, is already up and running.

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

st: Is a multilevel model appropriate?

From   Jennifer Day <>
Subject   st: Is a multilevel model appropriate?
Date   Mon, 13 Dec 2010 16:38:52 +1100

Hi out there,

I have district-level data on a country, and am interested in the
effects of metropolitan size on economic growth for districts that are
members of a metro region.  Metropolitan regions contain more than one
district.  Thus, I am interested in the effect on a district's growth,
of being a member of a large, medium, small, etc, metropolitan region.

My question is, is this an appropriate use of a mixed-level model?
There really is no random effect, as the district-level observations
are not samples of a larger population.  Are my standard errors biased
by the various levels if I use OLS, or is any part of BLUE
compromised?  Is there an alternate model that is appropriate?

The regressions are structured like this:

Growth in GRDP per capita (2001-04) = f[ initial GRDP,
infrastructure indicators from 2001,  human capital indicators from
2001,  natural endowment indicators from 2001,  capital indicators
from 2001, size of metropolitan agglomeration in 2001].

*   For searches and help try:

© Copyright 1996–2015 StataCorp LP   |   Terms of use   |   Privacy   |   Contact us   |   Site index