Speaker
Description
We address the problem of estimating the infection rate of an epidemic from observed counts of the number of susceptible, infected and recovered individuals. In our setup, a classical SIR (susceptible/infected/recovered) process spreads on a two-layer random network, where the first layer consists of small complete graphs representing the households, while the second layer models the contacts outside the households by a random graph. Our choice for the latter is the polynomial model, where three parameters control how the new vertices are connected to the existing ones: uniformly, preferentially, or by forming random triangles.
Our aim is to estimate the infection rate $\tau$. We apply two different approaches: the classical method uses a formula based on the maximum likelihood estimate, where the main information comes from the estimated number of the SI edges. The second, machine learning-based approach uses a fine-tuned XGBoost algorithm. We examine by simulations, how the performance of our estimators depend on the value of $\tau$ itself, the phase of the epidemic, and the graph parameters, as well as on the possible availability of further information.
Acknowledgement: This research has been implemented with the support provided by the Ministry of Innovation and Technology of Hungary from the National Research, Development and Innovation Fund, financed under the ELTE TKP 2021-NKTA-62 funding scheme.
| Classification | Both methodology and application |
|---|---|
| Keywords | SIR model, two layer random graph, xgboost |