README.md

In nikosbosse/epipredictr: What the Package Does (One Line, Title Case)

Epipredictr Workplan

To do

• [ ] Make model fit iteratively in the right way
• [ ] Write Text
• [ ] Do model averaging
• [ ] Reread Scoring rules

BSTS Models

• [x] Local
• [x] Local With Student Distribution
• [x] Semilocal
• [x] AR1
• [x] AR2
• [ ] STAN models? other models?
• [ ] Model Stacking / Averaging

Graphical Model Assessment

• [x] plot with predictions against true values
• [x] Plots of scoring across countries

Analytical Model Assessment and Scoring Rules

• [x] CRPS (use scoringRules package, should be straight-forward
• [x] Maybe LogS. Less straight forward and probably inferior, but also implemented in scoringRules

Popular examples of proper scoring rules for Ω = R include the logarithmic score and the continuous ranked probability score. The logarithmic score (LogS; Good 1952) is defined as LogS(F, y) = −log(f(y)),

Literature on that:

The packages caret (Kuhn et al. 2018) and forecast (Hyndman and Khandakar 2008) provide cross-validation tools suitable for cross-sectional and time series data, respectively. The loo(Vehtari et al. 2018) package implements recent proposals to select among Bayesian models. The ensembleBMA (Fraley et al. 2018) and ensembleMOS (Yuen et al. 2018) packages contain formulas for the CRPS of a small subset of the distributions listed in Table1 which are relevant for post-processing ensemble weather forecasts (Fraleyet al.2011), and can onlybe applied to specific data structures utilized in the packages. The surveillance (Meyer et al. 2017) package provides functions to compute the logarithmic score and other scoring rules forcount data models in epidemiology. The scoring (Merkle and Steyvers 2013) package focuses on discrete (categorical) outcomes, for which it offers a large number of proper scoring rules.

(Maybe future extensions)

• autoregressive time model
• semi-mechanistic model
• maybe machine learning
• maybe deep learning

nikosbosse/epipredictr documentation built on April 7, 2020, 9:51 p.m.