Table 1 Prediction targets.

From: Short-term local predictions of COVID-19 in the United Kingdom using dynamic supervised machine learning algorithms

Outcome

Mathematic formula

Y1pt: 1-week-ahead change in COVID-19 cases (as 1-week growth rate) compared with week t, by publication datea

\({\log }\frac{{{{{{\rm{Case}}}}}}{{P}}_{t+1}}{{{{{{\rm{Case}}}}}}{{P}}_{t}}\)

Y2pt: 2-week-ahead change in COVID-19 cases (as 2-week growth rate) compared with week t, by publication date

\({\log }\frac{{{{{{\rm{Case}}}}}}{{P}}_{t+2}}{{{{{{\rm{Case}}}}}}{{P}}_{t}}\)

Y3pt: 3-week-ahead change in COVID-19 cases (as 3-week growth rate) compared with week t, by publication date

\({\log }\frac{{{{{{\rm{Case}}}}}}{{P}}_{t+3}}{{{{{{\rm{Case}}}}}}{{P}}_{t}}\)

Y1st: 1-week-ahead change in COVID-19 cases (as 1-week growth rate) compared with week t, by collection date of specimenb

\({\log }\frac{{{{{{\rm{Case}}}}}}{{S}}_{t+1}}{{{{{{\rm{Case}}}}}}{{S}}_{t}}\)

Y2st: 2-week-ahead change in COVID-19 cases (as 2-week growth rate) compared with week t, by collection date of specimen

\({\log }\frac{{{{{{\rm{Case}}}}}}{{S}}_{t+2}}{{{{{{\rm{Case}}}}}}{{S}}_{t}}\)

Y3st: 3-week-ahead change in COVID-19 cases (as 3-week growth rate) compared with week t, by collection date of specimen

\({\log }\frac{{{{{{\rm{Case}}}}}}{{S}}_{t+3}}{{{{{{\rm{Case}}}}}}{{S}}_{t}}\)

  1. CasePt – number of COVID-19 cases by publication date at week t; CaseSt – number of COVID-19 cases by collection date of specimen at week t.
  2. aPublication date refers to the date when the case was registered on the reporting system.
  3. bCollection date of specimen refers to the date when the respiratory specimen was taken for testing.
Back to article page