Session timetable

Day 1: Delays and estimating the effective reproduction number

Monday June 23

Session 0: Introduction and course overview

Monday June 23: 09.00-09.30

Session 1: R, Stan, and statistical concept background

Monday June 23: 09.30-10.30

  • Introduction: statistical concepts used in the course (15 mins)
  • Introduction: stan concepts used in the course (15 mins)
  • Practice: introduction to estimation in stan (30 mins)

Session 2: Delay distributions

Monday June 23: 11.00-11.45

  • Introduction: epidemiological delays and how to represent them with probability distributions (10 mins)
  • Practice: simulate and estimate epidemiological delays (30 mins)
  • Wrap up (5 mins)

Session 3: Biases in delay distributions

Monday June 23: 11.45-12.30 and 14.00-14.45

  • Introduction: biases in delay distributions (10 mins)
  • Practice:
    • simulating biases in delay distributions and estimating delays without adjustment on these data (35 mins)
    • estimating delay distributions with adjustments for bias (35 mins)
  • Wrap up (10 mins)

Session 4: Using delay distributions to model the data generating process of an epidemic

Monday June 23: 14.45-15.30

  • Introduction: Using delay distributions to model the data generating process of an epidemic (15 mins)
  • Practice: implementing a convolution model and identifying potential problems (30 mins)

Session 5: \(R_t\) estimation and the renewal equation

Monday June 23: 16.00-17.30

  • Introduction: the time-varying reproduction number (10 mins)
  • Practice:
    • using the renewal equation to estimate R (35 mins)
    • combining \(R_t\) estimation with delay distribution convolutions (35 mins)
  • Wrap up (10 mins)

Day 2: An introduction to nowcasting and forecasting

Tuesday June 24

Tuesday June 24: 09.00-09.15

  • Day 1 review (15 mins)

Session 6: Nowcasting concepts

Tuesday June 24: 09.15-10.30

  • Introduction: nowcasting as a right-truncation problem (10 mins)
  • Practice:
    • simulating the delay distribution (25 mins)
    • nowcasting using pre-estimated delay distributions (30 mins)
  • Wrap up (10 mins)

Session 7: Nowcasting with an unknown reporting delay

Tuesday June 24: 11.00-12.30

  • Introduction: joint estimation of delays and nowcasts (10 mins)
  • Practice:
    • joint estimation of delays and nowcasts (35 mins)
    • joint estimation of delays, nowcasts and reproduction numbers (35 mins)
  • Wrap up (10 mins)

Session 8: Visualising infectious disease forecasts

Tuesday June 24: 14.00-14.45

  • Introduction: forecasting as an epidemiological problem, and its relationship with nowcasting and \(R_t\) estimation (10 mins)
  • Practice: extending a model into the future and visualising your forecast (30 minutes)
  • Wrap up (5 minutes)

Session 9: Forecast evaluation

Tuesday June 24: 14.45-15.30

  • Introduction: quantitatively evaluating forecasts (10 mins)
  • Practice: evaluating forecasts with a range of metrics (30 mins)
  • Wrap up (5 mins)

Session 10: Forecasting models

Tuesday June 24: 16.00-17.30

  • Introduction: a spectrum of forecasting models (10 mins)
  • Practice: evaluating forecasts from a range of models (60 mins)
  • Wrap up & discussion (20 mins)

Day 3: Forecast ensembles, methods in the real world and course summary

Wednesday June 25

Wednesday June 25: 09.00-09.15

  • Day 2 review (15 mins)

Session 11: Forecasting with ensembles

Wednesday June 25: 09.15-10.30

  • Introduction: strategies for collating and combining models (10 mins)
  • Practice: evaluating methods for ensemble forecasts (55 mins)
  • Wrap up (10 mins)

Session 12: Methods in the real world

Wednesday June 25: 11.00-12.00

  • Presentations and Q&A on uses of nowcasts & forecasts in the real world (60 mins)

Session 13: End of course summary

Wednesday June 25: 12.00-12.30