We then use synthetic data to test the performance in recovering a wide range of possible incidence trajectories.
The exact synthetic data used to test the algorithm was generated using the discrete multiplicative model data {N(t i )} of Eq. (20) with the parameter values α=6, β=1.5, a=4, b=2 and N 0=100.
First, we investigated the difference in power between an LR and a score test, finding that, although on synthetic data, the score test outperformed the LR test when effect sizes were small, the LR test found more associations when effect sizes were larger, as it did also on real data.
In our real data experiments (see below, and Section 4), we found that h 2 was typically on the order of a few percent (in the regime where, on synthetic data, the LR test tends to outperform the score test).
We validate our new approach with experiments on synthetic data, and we test and compare to related approaches its computational efficiency in data-intensive image segmentation applications.
Some areas are plotted as obstacles, in these datasets number of discrete obstacles have been drawn to consider obstacles in synthetic data sets for Test-data-110 data set (third row of Table 1) and the number of continuous and discrete obstacles have been drawn for Test-data-200 data set (fourth row of Table 1).
As in the regression experiment, 10-fold validation was used to estimate performance over real datasets and for synthetic data an independent test-set of size n=5000 was used.
In our first set of experiments, we conducted systematic experimentation using synthetic data in order to test the performance of the proposed methods of group graphical Granger modeling, primarily against that of the non-group variants.
On synthetic data, we examine two questions.
Using synthetic data generated by simulations, they tested the feasibility of such prediction.
