Figure 5

Methods and Models. Exposure effect’s estimation according to Methods 1 and 2 and related to: (A) The HP model, (B) The L W A _u model, (C) The L W A _a model and (D) The L W A _i model. The solid thick lines red, blue and green (Figure A, B and C) respectively, represent the theoretical average of $\gamma \left(t\right)$ and the solid thick orange lines (Figure D) represent the theoretical $\gamma \left(t\right)+{\mathbf{\alpha }}^{{}^{\prime }}\mathbf{\left(}\mathit{t}\mathbf{\right)}\mathit{Z}$ inside each profile Z. The fine lines red, blue and green (Figure A, B and C) respectively, represent $\gamma \left(t\right)$, estimated according to Methods 1 (fine dotted lines) and 2 (solid dotted lines); and the fine red lines (Figure D) represent the observed $\gamma \left(t\right)+{\mathbf{\alpha }}^{{}^{\prime }}\mathbf{\left(}\mathit{t}\mathbf{\right)}\mathit{Z}$ inside each profile. The average bias and RMSE are given for each model: b _l•, b _••, R M S E _l• and R M S E _••. These results are obtained with the increasing then decreasing H R(t) configuration, without censoring and with ${\mathit{\beta }}_{12}^{{}^{\prime }}=(-0.2,-0.4,-0.8)$, ${\mathit{\beta }}_{13}=-{\mathit{\beta }}_{12}$ and ${\mathit{\beta }}_{23}=-{\mathit{\beta }}_{13}$.