Agriculture Reference
In-Depth Information
Table 5.2 Results obtained with time series econometric techniques, considering the Cobb-
Douglas (1928) model extended with economic, social, and environmental variables (linear
model obtained with logarithms), for the agricultural output in the period 1990-2011 (there are
not results for Greece, due to a lack of data)
Spain
France
Ireland
Italy
UK
Prais-
Winsten
(Robust)
Prais-
Winsten
(Robust)
Prais-
Winsten
Prais-
Winsten
Prais-
Winsten
Model
Constant
28.013*
( 3.380)
[0.004]
6.129*
(3.080)
[0.007]
11.334*
( 3.900)
[0.001]
18.159*
( 2.570)
[0.020]
6.061*
(3.530)
[0.002]
Agriculture value added per
worker (constant 2005 US$)
0.813*
(7.060)
[0.000]
0.744*
( 4.020)
[0.001]
Employment in agriculture
(% of total employment)
0.926*
(4.740)
[0.000]
2.929*
(8.590)
[0.000]
0.813*
(2.500)
[0.023]
0.747*
(3.280)
[0.004]
Additional variable a
6.134*
(2.350)
[0.031]
0.826*
(3.580)
[0.003]
1.540*
(2.960)
[0.009]
3.374*
(2.670)
[0.016]
0.617*
(
3.850)
[0.001]
Breusch-Pagan/Cook-
Weisberg test for
heteroskedasticity
0.010
[0.931]
1.360
[0.244]
3.000**
[0.083]
0.240
[0.621]
4.180*
[0.040]
Ramsey RESET test using
powers of the fitted values
2.190
[0.135]
0.160
[0.919]
2.060
[0.151]
3.860*
[0.033]
2.110
[0.141]
LM test for autoregressive
conditional heteroske-
dasticity (ARCH)
0.065
[0.798]
0.717
[0.397]
1.411
[0.234]
0.800
[0.371]
0.097
[0.755]
Note : *Statistically significant at 5 %; **Statistically significant at 10 %
a
Rural population (% of total population) for Spain, Exports of goods and services (% of GDP) for
France and Ireland, Fossil fuel energy consumption (% of total) for Italy, and GDP per capita
(current US$) for the UK
the percentage of the gross capital formation and did not present statistical signif-
icance). In Table 5.2 the results presented for the models (countries) where the
Ramsey RESET test using powers of the fitted values reveal a lack of independent
variables. In this case several economic, social, and environmental variables ana-
lyzed in the previous section were tested, taking into account the availability of
statistical information considered, in all estimations, only the period from 1990 to
2011. All the econometric estimations were made with Stata ( 2014 ) software.
Observing Table 5.1 it is possible to verify that there are no problems with the
unit root and with the co-integration. There are some problems, however, with the
autocorrelation and because of this the Prais-Winsten was used as an estimation
method. On the other hand there are some complications with the heteroskedasticity
and with the autoregressive conditional heteroskedasticity, for the model of the UK,
and in this way the robust ARCH family regression was considered for the estima-
tion. The Ramsey RESET test using powers of the fitted values reveal that there is a
lack of independent variables in the models of Spain, France, Ireland, Italy, and the
Search WWH ::




Custom Search