Selected article for: "APE metric and predictive performance"

Author: KV Parag; CA Donnelly
Title: Optimising Renewal Models for Real-Time Epidemic Prediction and Estimation
  • Document date: 2019_11_8
  • ID: 9t5ncsig_22
    Snippet: When computing directly, the most difficult term is B s+1 when I s+1 and α τ (s) 176 are large. In these cases Stirling approximations may be applied. The renewal 177 model APE metric offers a simple means of finding k * := arg min k APE k , the 178 window length that optimises one-step-ahead predictive performance......
    Document: When computing directly, the most difficult term is B s+1 when I s+1 and α τ (s) 176 are large. In these cases Stirling approximations may be applied. The renewal 177 model APE metric offers a simple means of finding k * := arg min k APE k , the 178 window length that optimises one-step-ahead predictive performance.

    Search related documents: