What’s the pandemic forecast?
What’s the pandemic forecast?
What’s going to happen next with the COVID-19 pandemic in Australia? What about a few weeks from now? What about in different states and territories?
These are the questions a team of researchers from the ARC Centre of Excellence for Mathematical and Statistical Frontiers (ACEMS) is helping to answer.
Since March, the group has been busy modelling the pandemic along with other researchers across the nation. Every week, the results of that modelling shows up on the desks of Australia's top leaders who are making critical decisions about things like social distancing, whether to reopen economies, or keep people at home.
“It’s an interesting time to be producing real-time forecasts that turn into policy within a few days,” said Professor Rob Hyndman, Head of the Department of Econometrics and Business Statistics at Monash University and an ACEMS Chief Investigator.
For years, Professor Hyndman has helped companies and government agencies around the world forecast things like product sales, energy usage, and tourism. But not epidemics.
“Forecasters and epidemiologists haven’t worked together much before. Epidemiologists usually study disease dynamics and measures to reduce transmission. But with a pandemic occurring, we needed to join forces in order to provide the best advice on how to manage the situation,” Professor Hyndman said.
Professor Hyndman’s primary role is to produce an ensemble forecast from several different mathematical models on the spread of COVID-19 here in Australia.
“An ensemble is when you combine forecasts from multiple models to get an overall forecast. If one model leaves out some important information, it hopefully gets picked up in one of the other models. Generally, you get better forecasts than if you took any of the models on their own.”
“Some of this is cutting-edge research. Part of what we're trying to do with the ensemble looks at the quality of each of the models each week and then adjust the weights to favour the good models. There’s not much available research in that space, so it’s interesting and hopefully useful,” Professor Hyndman said.
The team’s efforts and learnings will continue for months to come, as leaders turn to the mathematical sciences to provide critical information on what may be coming next.
Image: Professor Rob Hyndman. Credit: Monash University.