The first Data Science seminar of this term will take place on Wednesday 29/9, from 15:00 to 16:00, in amphitheatre LRC012 – “Stelios Ioannou” Learning Resource Center.
The speaker will be Dr Xenia Miscouridou who is currently a postdoctoral researcher at Imperial College London, and the title of her talk will be
“Predicting the present: An application on Covid-19 mortality”
Abstract: Updating observations of a signal due to the delays in the measurement process is a common problem in signal processing, with prominent examples in a wide range of fields. An important example of this problem is the nowcasting of COVID-19 mortality: given a stream of reported counts of daily deaths, can we correct for the delays in reporting to paint an accurate picture of the present, with uncertainty? Without this correction, raw data will often mislead by suggesting an improving situation. We present a flexible approach using Gaussian processes that can describe the changing auto-correlation structure in the reporting time-delays. This approach yields robust estimates of uncertainty for the estimated numbers of deaths and performs favourably against both comparable methods, and against a small sample of expert human predictions. When tested on a real dataset of COVID-19 mortality data from Brazil with reporting delays are large, our approach is able to make informative predictions on important epidemiological quantities such as the current effective reproduction number.