INFORMATIVE PRIOR DIFFUSE PRIOR
|
NORTH |
CENTER |
SOUTH |
NORTH |
CENTER |
SOUTH | |
|
σ###KL |
0.13841 |
0.05088 |
0.03848 |
-0.11992 |
-0.24101 |
-0.31822 |
|
st. err. |
0.11212 |
0.08993 |
0.10634 |
0.35575 |
0.31708 |
0.23709 |
|
^##KK |
-0.27205 |
-0.10575 |
-0.07020 |
0.23032 |
0.50949 |
0.58942 |
|
st. err. |
0.22055 |
0.18890 |
0.19623 |
0.67920 |
0.67211 |
0.44091 |
|
σ≡LΓ- |
-0.07054 |
-0.02450 |
-0.02112 |
0.06250 |
0.11416 |
0.17195 |
|
st. err. |
0.05737 |
0.04297 |
0.05783 |
0.18763 |
0.15064 |
0.12808 |
|
ε≡KΓ^ |
0.09196 |
0.03438 |
0.02472 |
-0.07884 |
-0.16357 |
-0.20661 |
|
st. err. |
0.07438 |
0.06050 |
0.06929 |
0.23317 |
0.21520 |
0.15408 |
|
ε≡LK- |
0.04689 |
0.01655 |
0.01356 |
-0.04108 |
-0.07744 |
-0.11161 |
|
st. err. |
0.03798 |
0.02889 |
0.03764 |
0.12271 |
0.10201 |
0.08310 |
|
£###KK |
-0.09196 |
-0.03438 |
-0.02472 |
0.07884 |
0.16356 |
0.20661 |
|
st. err. |
0.07438 |
0.06050 |
0.06929 |
0.23317 |
0.21520 |
0.15408 |
|
ε≡LΓ- |
-0.04689 |
-0.01655 |
-0.01356 |
0.04107 |
0.07744 |
0.11161 |
|
st. err. |
0.03798 |
0.02889 |
0.03764 |
0.12271 |
0.10201 |
0.08310 |
TABLE 3 - Elasticities conditional on 1989 (Y,t,w/r) without inequality constraints
A few comments are worthwhile. As shown in Table 2 and 3, expected values
conditional on labor share and on (Y,t,w/r) without imposing concavity are very similar.
Therefore we can conclude that the standard translog model predict quite well in terms
of elasticities. Very different is the situation when we impose concavity. Results for
expected values and standard deviations are tabulated in Table 4.
14