||Predicting Structural Changes in Danish Agriculture
||Christiansen, Lasse Engbo (Mathematical Statistics, Department of Informatics and Mathematical Modeling, Technical University of Denmark, DTU, DK-2800 Kgs. Lyngby, Denmark)
||Technical University of Denmark, DTU, DK-2800 Kgs. Lyngby, Denmark
||In this thesis, different models are built to examine the structural change of the
pig farms in Denmark and predict the farm distribution in 2030.
At first, aggregate data are used in the modelling with only the number of
farms in each size category from 1982 to 2009. Regression models are estimated
and and make predictions, but the results do not accord with the practical
experience. Then the Markov model is estimated using both generalised cross
entropy (GCE) method and the ordinary least squares (OLS) method. The
GCE estimator may give the results not as precise as the OLS, but when the
data points are limited the GCE can always provides a robust estimation, but
the OLS may have a large variance. And the predictions made by the Markov
model are more plausile than the result from the regression model.
Then the CHR data are used to estimate the transition probability matrix
(TPM) in the Markov model. And both the time variation and regional effect of the estimated TPM are checked. It is concluded that the TPM can be
supposed to be stationary while the TPM on the farms from different regions
may vary. The estimated TPM are use to predict the farm distribution in 2030,
and a restructuring of the pig farms are detected. Finally the spatial distribution of the pig farms are displayed to see the areal concentration of the pig
farms of certain types.
||Technical University of Denmark (DTU) : Kgs. Lyngby, Denmark
Creation date: 2011-02-03
Update date: 2011-02-03