I am looking for papers, preferably in economics, where simple OLS is used. The reason is that to convey the workings of linear regression, I want to have real data from real papers. Most research uses simultaneous models or other bells and whistles. Also few researchers release their data. I could generate my own data, but as I said, I want to have the whole story of the data and research behind it. Wooldridge (Introductory Econometrics) does a lot of simple analyses from past papers, but as far as I know, these are just simplified version of the real models.
3 Answers
$\begingroup$
$\endgroup$
0
Hayashi is a good book to start with. http://press.princeton.edu/titles/6946.html
To be more specific you will find the paper of Returns to Scale in Electricity Supply by Marc Nerlove in section 1.7. It is a simple application of OLS. Yet it give a good understanding of OLS.
You will find the full paper and the data here. There are further details in the book. You can download the first chapter of the book for free. The link is on the same website.
$\begingroup$
$\endgroup$
0
One very famous paper that uses simple OLS that any 1st year economics undergraduate can understand and recreate is:
Mankiw, N. Gregory, David Romer, and David N. Weil (1992), “A Contribution to the Empirics of Economic Growth.” The Quarterly Journal of Economics 107: 407-437.
It's a great early contribution to the growth literature, and a very easy read! It uses data from the Summer and Heston World Tables which are freely available!
A Google will reveal .pdf copies of this paper in various places.
Also, let me know if you're not too into macro/growth and want something different. I can try to think up some cooler papers that only use OLS.
$\begingroup$
$\endgroup$
The Statistical Sleuth by Ramsey and Schafer
http://www.science.oregonstate.edu/~schafer/Sleuth/
is especially good at digging deep into the research goals and context of particular papers with as far as possible real and original data. It extends across several disciplines and covers regression-type methods in depth, and some others too. So, it should be of interest to readers who do not want to specify "preferably in economics".