Skip to main content
. 2017 Jul 20;13(7):e1005607. doi: 10.1371/journal.pcbi.1005607

Table 1. Comparison of ARGO to benchmark models across countries and evaluation metrics.

The bold face value is the best value among all methods according to each performance metric. Google Dengue Trends was not published for Taiwan and therefore the GDT benchmark is not available for Taiwan. The assessment period for the five regions, chosen based on the common available periods for all methods, are: Brazil (Mar 2006–Dec 2012), Mexico (Mar 2006–Aug 2015), Thailand (Oct 2010–Aug 2015), Singapore (Feb 2008–Aug 2015), Taiwan (Jan 2013–Mar 2016). The error value is relative to the naive, whose absolute error value is reported in the parenthesis.

RMSE MAE RMSPE MAPE CORR
Brazil
ARGO 0.394 0.369 0.397 0.389 0.971
GDT 0.666 0.633 0.984 0.817 0.916
GT 0.902 0.829 0.877 0.838 0.861
SAR 0.660 0.563 0.664 0.583 0.917
SAR+GDT 0.629 0.587 0.564 0.560 0.938
naive 1 (30560.436) 1 (21677.634) 1 (0.703) 1 (0.546) 0.812
Mexico
ARGO 0.680 0.651 0.558 0.678 0.924
GDT 0.944 0.961 1.270 1.311 0.863
GT 0.950 0.927 1.097 1.100 0.861
SAR 0.790 0.737 0.776 0.815 0.911
SAR+GDT 1.249 0.986 0.779 0.854 0.891
naive 1 (3570.105) 1 (2161.018) 1 (0.816) 1 (0.492) 0.833
Thailand
ARGO 0.715 0.715 0.708 0.706 0.928
GDT 0.880 0.868 1.494 1.284 0.884
GT 1.364 1.224 1.510 1.368 0.833
SAR 0.774 0.836 0.906 0.898 0.917
SAR+GDT 1.157 0.983 0.923 0.936 0.903
naive 1 (2058.891) 1 (1276.068) 1 (0.426) 1 (0.326) 0.852
Singapore
ARGO 0.893 0.889 0.931 0.917 0.903
GDT 1.182 1.285 1.427 1.439 0.821
GT 1.287 1.165 1.287 1.254 0.796
SAR 1.153 1.104 1.166 1.087 0.847
SAR+GDT 2.452 1.297 1.185 1.009 0.775
naive 1 (329.318) 1 (202.651) 1 (0.283) 1 (0.230) 0.878
Taiwan
ARGO 2.180 1.264 0.233 0.359 0.834
GT 12.211 4.904 1.069 0.898 0.724
SAR 1.852 1.397 0.247 0.408 0.878
naive 1 (2422.559) 1 (1063.597) 1 (3.248) 1 (1.601) 0.734