Q= αLβ1Kβ2
We take log to make the model loglinear and make it suitable for OLS regression. The regression output is given by:
SUMMARY OUTPUT
Regression Statistics
Multiple R 0.973748
R Square 0.948186
Adjusted R Square 0.93955
Standard Error 0.038937
Observations 15
ANOVA
significance
df SS MS F F
Regression 2 0.332922 0.166461 109.7978 1.94E-08
Residual 12 0.018193 0.001516
Total 14 0.351115
Standard
Coefficients Error t Stat P-value Lower 95%
Intercept -2.06496 0.349946 -5.90079 7.24E-05 -2.82742
logK 0.415207 0.134517 3.086656 0.009421 0.12212
Log L 1.078004 0.249335 4.323522 0.00099 0.53475
The estimated model is given by:
Log Q==2.06+.41log K+1.07logL
Here, p value of log K&L are<.01. thus, the coefficient of capital and labor are statistically significant at 1% level.
2. Here, p value of log K&L are<.01. thus, the coefficient of capital and labor are statistically significant at 1% level.
3 . Here, R^2=.94, thus 94 percentage of the variation in output is explained by the regression equation
4 .
EK = β1 = 0.515 ( 1% increase in K yield a 0.415% increase in Q)
EL = β2 =1.078 ( 1% increase in K yield a 1.078% increase in Q)
5 .
Since the sum of the exponents of the capital and labor inputs exceeds 1.0 (1.493), the production function exhibits increasing returns to scale.
Comments
Leave a comment