Fig. 3
From: Forecasting Chikungunya spread in the Americas via data-driven empirical approaches
Comparison of observed versus predicted Chikungunya cases in Curacao and the United States. Incomplete, intermittent, and delay in reports generated inaccurate model calibration with consequent incorrect predictions. Top: United States. Consistent patterns of report submission allowed us to anticipate imported cases with numbers predicted close to real numbers of cases. Bottom: Curacao. While observed reports (red line) showed low increases in the first two weeks of the outbreak, the country was characterized by dramatic increases of cases with irregular reporting accumulating numbers of cases such that we could not generate correct forecasts for the following six (purple), five (blue), four (dark green), and three months (light green)