Selected article for: "accurate model and actual outcome"
Author: Sarah F. McGough; Michael A. Johansson; Marc Lipsitch; Nicolas A. Menzies
Title: Nowcasting by Bayesian Smoothing: A flexible, generalizable model for real-time epidemic tracking
  • Document date: 2019_6_7
    • ID: 6kq0ptlg_29
    Snippet: The copyright holder for this preprint (which was not peer-reviewed) is the author/funder. . https://doi.org/10.1101/663823 doi: bioRxiv preprint both models could produce forecasts (weeks with at least one case initially reported), point estimates for NobBS were substantially more accurate than the benchmark model for dengue cases (rRMSE improved by 300%) and slightly less accurate for ILI cases (rRMSE decreased by 19%). However, analysis of the.....
    Document: The copyright holder for this preprint (which was not peer-reviewed) is the author/funder. . https://doi.org/10.1101/663823 doi: bioRxiv preprint both models could produce forecasts (weeks with at least one case initially reported), point estimates for NobBS were substantially more accurate than the benchmark model for dengue cases (rRMSE improved by 300%) and slightly less accurate for ILI cases (rRMSE decreased by 19%). However, analysis of the probability distributions of the nowcasts revealed a much more substantial difference; the average score for NobBS was approximately twice as high for dengue and more than 10 times as high for ILI cases (Table 1 ). This indicates that the NobBS approach assigned much higher probability to the actual outcome, even though point accuracy was somewhat lower for the ILI cases.

    Search related documents:
    Co phrase search for related documents
    • average score and high probability: 1, 2
    • average score and NobBS approach: 1, 2
    • average score and NobBS average score: 1, 2
    • average score and point accuracy: 1
    • benchmark model and dengue case: 1
    • benchmark model and NobBS approach: 1, 2, 3
    • benchmark model and NobBS average score: 1
    • benchmark model and point accuracy: 1, 2
    • benchmark model and point estimate: 1
    • benchmark model and probability distribution: 1
    • dengue case and ILI case: 1, 2
    • dengue case and NobBS approach: 1, 2, 3
    • dengue case and point accuracy: 1, 2, 3
    • dengue case and point estimate: 1
    • forecast produce and model forecast produce: 1, 2
    • high probability and point estimate: 1
    • ILI case and point accuracy: 1
    • ILI case and point estimate: 1
    • NobBS approach and point estimate: 1