Transcript PowerPoint


Problem Setup
› First, we set up a production function with
four inputs
1
f  x    x  xAx
2
  10 15 5 1.5
 0.75 0.06 0.10 0.02 
 0.06 0.35 0.12 0.008 

A
 0.10 0.12 0.50 0.014 


 0.02 0.008 0.014 0.10 
› Inputs 1, 2, and 3 are variable inputs whose
levels are determined by relative prices.
› Input 4 is the quasi-fixed input (capital)
which yields the individual effect (both
random and fixed).
› We assume initially that input 4 is
unobservable.

In this sample, we draw 1200
observations.
› 40 individuals (N = 40)
› 30 time periods (T = 30)

Start by estimating the Covariance
estimator as we did for the fixed effects
model (or the within estimator).
ˆCV
1
40






   X i QX i    X i Qyi 
 i 1
  i 1

40

Next, we estimate the between
estimator
1
40





     xi  x  xi  x      xi  x  yi  y  
 i 1
  i 1

40