×

zbMATH — the first resource for mathematics

Parameter estimation and uncertainty quantification for an epidemic model. (English) Zbl 1260.62085
Summary: We examine estimation of the parameters of Susceptible-Infective-Recovered (SIR) models in the context of least squares. We review the use of asymptotic statistical theory and sensitivity analysis to obtain measures of uncertainty for estimates of the model parameters and the basic reproductive number (\(R_0\)), an epidemiologically significant parameter grouping. We find that estimates of different parameters, such as the transmission parameter and recovery rate, are correlated, with the magnitude and sign of this correlation depending on the value of \(R_0\). Situations are highlighted in which this correlation allows \(R_0\) to be estimated with greater ease than its constituent parameters. Implications of correlations for parameter identifiability are discussed. Uncertainty estimates and sensitivity analysis are used to investigate how the frequency at which data is sampled affects the estimation process and how the accuracy and uncertainty of estimates improves as data is collected over the course of an outbreak. We assess the informativeness of individual data points in a given time series to determine when more frequent sampling (if possible) would prove to be most beneficial to the estimation process. This technique can be used to design data sampling schemes in more general contexts.

MSC:
62P10 Applications of statistics to biology and medical sciences; meta analysis
92D30 Epidemiology
62F12 Asymptotic properties of parametric estimators
92C60 Medical epidemiology
62M10 Time series, auto-correlation, regression, etc. in statistics (GARCH)
62H20 Measures of association (correlation, canonical correlation, etc.)
PDF BibTeX XML Cite
Full Text: DOI