Reform of the EU Sugar Regime: Impacts on Sugar Production in Ireland

Table 2: Expected Relationship between variables

Model Estimation

To estimate the relationship between these variables and relative burdens
for individual countries, the paper uses pooled time-series cross-section
(TSCS) ordinary least square regressions (Stimson 1985) with panel
corrected standard errors (PCSEs) (Beck and Katz 1995).²⁸ Prais-Winston
transformations are used to eliminate serial correlation of the errors and
to take account of cross-section and panel specific auto-correlation. In
running the regression of the above independent variables on the number
of relative asylum seekers, I lagged GDP growth, unemployment, foreign
population and the deterrence index by one year as one might reasonably

28 'Pooled', 'panel' or 'TSCS' analysis has become a popular tool for the empirical analysis
of issues in Comparative Politics and International Relations. It involves the analysis of
N cross-sections (countries) and T time periods (years). It increases the number of
observations available and allows for the analysis of dynamic factors in cross-national
comparative research. The paper corrects for expected downward bias in standard errors
and upward bias in t-statistics (Hicks 1994) by eliminating serial correlation of the errors
applying Beck and Katz's standard method of 'panel corrected standard errors' (PCSE).
For a recent review on 'pooling' see Beck (2001).

22

<<   <   21   22   23   24   25   26   27   >   >>