Apart from conventional model performance metrics, a new index, called percent volumetric error, was proposed, especially for quantifying the phase error.
Three model performance metrics were calculated: Nash Sutcliffe Efficiency, a normalised error index statistic (the ratio of the root mean square error to the standard deviation of observed data) and the percentage bias.
In the review process, network parameters, distance, speed, no. of nodes, no. of hops, communication overhead, relative mobility, simulation parameters, simulation tool used, mobility model, performance metrics and compared protocols are taken under consideration.
Overall, improved simulations were noted for adjusted-LBC runs for ozone concentrations both at the ground-level and in the middle to upper troposphere, based on EPA-recommended model performance metrics.
Results of this study show that the automatic calibration of such daily models will be misled by model performance metrics that are based on the aggregation of daily residuals to a solution that systematically underestimate the peak flow rates while trying to emulate the one-day lags.
In contrast, model performance metrics indicated that V 2 was a moderately good predictor of mean log MOE (RMSE 1.4 GPa, MAPE 11%%, TP rate 74 %, validation dataset).
This is followed by the explanation of the approximated analytical model for proposed scheme in Section 3. Sections 4 and 5 present simulation model, performance metrics and results from the simulations carried out respectively.
Based on the above, for each taxon, we report three model performance metrics: (1) the 25-fold average area under the ROC curve (AUC) [55], [61], [64], [66] of test data (ATAUC), (2) the standard deviation of the test AUC of the 25 different folds (STAUC), and (3) the proportion of the potential distribution coverage with standard deviation above 0.15 (ASD15).
The autologistic approach allows obtaining unbiased estimates of each model variable coefficient and significance, but it tends to artificially inflate the autologistic model performance metrics index because it incorporates information on the neighboring sites in the same data set.
To seek the optimal choice, different pre-processing strategies were evaluated and the most accurate was selected according to model performance metrics (i.e., lowest RMMCCV and r closet to 1) and confirmed by its high predictive accuracy (i.e., low RMSEP, high RPD and RER values) [ 41].
