Statistical Analysis with Bootstrap Diagnostics of Atmospheric Pollutants Predicted in the Apsis Experiment
- Cite this article as:
- Archer, G. & Giovannoni, J. Water, Air, & Soil Pollution (1998) 106: 43. doi:10.1023/A:1005004022883
- 37 Views
Predictions of O3 and NO2 delivered by three eulerian models the APSIS study are compared with observed levels of the phenomen recorded on 25 May 1990, over the greater Athens area, Greece. A variety of measures are used to test model performances, including normalised mean square error, bias and correlation, for which confidence intervals are consteucted. Causes of differences in the model differences are briefly discussed. A jackknife diagnostic check is performed on the adequacy of the bootstrap procedure used to make the intervals, which are found to be suitably robust. Similarities in model behaviour, and specific performancies, are detected. Some statistical criteria seem to give a slight advantage to one of the three sets of model results, although important biases can occur locally, between observations and predictions. On the whole, no one of the set of model results emerges as significantly superior to the others.