Figure 4.
True vs. fitted partial effect weight function h(t−tz) for scenario L=100 players and σb=0.5. Rows: Uncons. model (top row), Constr. model (middle row) and Ridge model (bottom row). Columns: (a) exponential decay, (b) bi-linear, (c) early peak and (d) inverted U. The true shapes used for simulation are depicted as solid black lines and the mean (point-wise averages) of all simulation runs are shown as solid coloured lines. A random sample of 100 individual estimated weight functions are displayed as light-coloured curves.

True vs. fitted partial effect weight function h(ttz) for scenario L=100 players and σb=0.5. Rows: Uncons. model (top row), Constr. model (middle row) and Ridge model (bottom row). Columns: (a) exponential decay, (b) bi-linear, (c) early peak and (d) inverted U. The true shapes used for simulation are depicted as solid black lines and the mean (point-wise averages) of all simulation runs are shown as solid coloured lines. A random sample of 100 individual estimated weight functions are displayed as light-coloured curves.

Close
This Feature Is Available To Subscribers Only

Sign In or Create an Account

Close

This PDF is available to Subscribers Only

View Article Abstract & Purchase Options

For full access to this pdf, sign in to an existing account, or purchase an annual subscription.

Close