Table 6 Comparison of MAEs on German health care demand dataset.
From: Robust learning for ridge-penalized quasi-GLMs under non-identical distributions
MAEs for NBR | |||||
|---|---|---|---|---|---|
Year | \(\beta\) | DOCVIS | HOSPVIS | ||
Non-truncation | Truncation | Non-truncation | Truncation | ||
1984 | 1.9 | 2.982 | 2.686 | 0.256 | 0.127 |
1985 | 1.7 | 2.869 | 2.688 | 0.299 | 0.156 |
1986 | 1.7 | 3.230 | 2.982 | 0.270 | 0.132 |
1987 | 1.9 | 3.022 | 2.840 | 0.304 | 0.148 |
1988 | 1.8 | 2.568 | 2.398 | 0.294 | 0.155 |
1991 | 2.0 | 2.350 | 2.167 | 0.286 | 0.149 |
1994 | 1.9 | 3.356 | 3.038 | 0.277 | 0.140 |
