So I am currently trying to write a paper estimating the Constant Elasticity of Substitution Production Function of the USA. I am using the simple version with two inputs capital and labour. Since the estimation is at the aggregate level I thought of using the perpetual inventory method to create the data for aggregate capital, but I wasnt sure what kind of data I need for aggregate labour. Maybe the number of employed people, work force, in the USA? I'd be glad if someone could help me out.


The Penn Wolrd Tables have the data you need.

| improve this answer | |
  • $\begingroup$ so I downloaded it from here and read through the manual rug.nl/ggdc/productivity/pwt so for L I would then take the data for emp (employed people) but since there is a high autocorrelation in there wouldnt that be a problem for the estimations further? Same for capital instead of me constructing the aggregate capital data can't I just take the column rkna (Capital stock at constant 2011 national prices (in mil. 2011US$), but again there is high autocorrelation involved, sorry if my questions are very beginnerish. $\endgroup$ – macro123 Feb 8 '18 at 23:23
  • $\begingroup$ @macro123 Your questions are to the point. But estimation in an autocorrelation framework is nothing new, you just have to read the relevant time-series econometric literature on how to handle it. $\endgroup$ – Alecos Papadopoulos Feb 8 '18 at 23:42

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.