| | Specification A | Specification B |
|---|
Age group |
\( { y_{ij,t-1} } \)
|
\( { u_{ij,t-1} } \)
| Joint |
\( { y_{ij,t-1} } \)
|
\( { u_{ij,t-1} } \)
| Joint |
|---|
Up to 18 | 1% | 3% | 4% | 0% | 19% | 19% |
18 to 25 | 29% | 21% | 50% | 19% | 8% | 27% |
25 to 30 | 18% | 14% | 31% | 54% | 11% | 65% |
30 to 50 | 1% | 5% | 6% | 5% | 8% | 13% |
50 to 65 | 1% | 1% | 1% | 2% | 0% | 2% |
Over 65 | 1% | 0% | 2% | 1% | 1% | 2% |
- Note: Specification A is based on the computation of the squared correlation of the respective regressor with the dependent variables (univariate \( { R^2 } \)). Specification B is calculated using the estimated SYS-GMM coefficient from the augmented migration model specification in Table 9 (see the Appendix). The_estimation coefficient for regressor \( { x_k } \) is further standardized as \( { {\hat{\beta}_{\text{standardized},{k}}}\,{=}\,{\hat{\beta}_{{k}}}\,{{\sqrt{{s}_{{kk}}}}/{\sqrt{{s}_{{yy}}}}} } \), where \( { {s}_{{kk}} } \) and \( { {s}_{{yy}} } \) denote the empirical variances of regressor \( { x_k } \) and the dependent variable \( { y } \), respectively. As long as one only compares regressors within models for the same \( { y } \), division by \( { \sqrt{{s}_{{yy}}} } \) is irrelevant.
