Table 4. Estimates of the input coefficients in the I-district fixed effects.
|
________________(1.1) |
(1.2) |
_____________(1.3) |
(1.4) |
(1.5) |
(1.6) | |
|
Constant |
5.1464 *** |
5.5055 *** |
5.5861*** |
4.7451*** |
4.7635*** |
4.8347*** |
|
(0.000) |
(0.000) |
(0.000) |
(0.000) |
(0.000) |
(0.000) | |
|
R&D firms (1) |
0.2635 *** |
0.2825 *** |
0.2751*** | |||
|
(0.000) |
(0.000) |
(0.000) | ||||
|
R&D firms (2) |
0.0905*** |
0.0970*** |
0.0943*** | |||
|
(0.000) |
(0.000) |
(0.000) | ||||
|
R&D public |
0.1902 *** |
0.2441 *** |
0.2485*** |
0.2061*** |
0.2328*** |
0.2187*** |
|
___________(0.001) |
(0.000) |
________(0.000) |
(0.001) |
(0.000) |
(0.000) | |
|
Territory and |
Territory and | |||||
|
Fixed effects |
Territory |
Industry |
Industry |
Territory |
Industry |
Industry |
|
Fixed effects F-test |
0.000 |
0.000 |
0.000 |
0.000 |
0.000 |
0.000 |
|
LR selection (Iambda=O)_________ |
_____________0.000 |
0.0001 |
________0.0002 |
0.0003 |
0.0003 |
0.0008 |
|
R2-ajd / Pseudo R2 |
0.297 |
0.3225 |
0.3283 |
0.3111 |
0.3315 |
0.3371 |
|
Log-L |
-681.46 |
-670.24 |
-658.20 |
-679.82 |
-666.19 |
-654.26 |
|
Akaike |
1370.91 |
1378.473 |
1390.40 |
1379.64 |
1370.39 |
1382.53 |
|
BIC |
1388.50 |
1462.015 |
1553.08 |
1423.61 |
1453.93 |
1545.22 |
|
Number of obs________________ |
_____________806 |
_______806 |
___________806 |
_______806 |
806 |
_________806 |
(1) R&D imputed from regional data
(2) R&D from firm microdata
Notes: (a) Dependent variable = Patents per employee in the 2001-2006 period; (b) All variables are natural logarithms; (c) P-values are in parentheses and
asterisks represent statistical significance at 1% (***), 5% (**) and 10% (*); (d) Within group effect model estimates; (e) Fixed effects provided under the
restriction that ∑ αi = 0, so that the dummy coefficients mean deviations from the averaged group effect (intercept); (f) Heckman two stages coefficients
adjusted for sample selection; (g) Robust Huber-White estimators when slight problems of heteroskedasticiy, collinearity or outliers are detected.
18
More intriguing information
1. The name is absent2. Effects of red light and loud noise on the rate at which monkeys sample the sensory environment
3. The Impact of Cognitive versus Affective Aspects on Consumer Usage of Financial Service Delivery Channels
4. The name is absent
5. The name is absent
6. EXECUTIVE SUMMARY
7. The name is absent
8. Text of a letter
9. The name is absent
10. Eigentumsrechtliche Dezentralisierung und institutioneller Wettbewerb