Nowcasting by Bayesian Smoothing: A flexible, generalizable model for real-time epidemic tracking

Sarah F. McGough; Michael A. Johansson; Marc Lipsitch; Nicolas A. Menzies

The performance of ILI nowcasts across accuracy measures was relatively consistent by year, but there were fluctuations in the year-to-year performance of both approaches applied to dengue data (Table 3) . Average scores tended to be high in years that experienced a very low number of dengue cases (e.g. 2000, 2002, 2004, 2006) . The model was particularly effective at identifying periods of low incidence, with high probabilities assigned to the lowest outcome bin (0-25 cases, details in Materials and Methods) when the number of cases eventually reported was low (Fig. S5) . On the other hand, during periods of high dengue activity, lower probabilities were assigned to the correct bin, reflecting greater uncertainty. Overall, NobBS outperformed the benchmark approach on all performance measures for each year (Table 3) . 0.17 across the NobBS and benchmark approaches, falling clearly below the rest of the years in performance. These scores not only reflect unusually poor point estimate predictions as judged by rRMSE, but also the finding that the predictive distribution for weeks in these years for both approaches rarely included the true value of interest (a consequence of dramatic over-or underestimates).