scholarly journals COVID-19 outbreak in Wuhan demonstrates the limitations of publicly available case numbers for epidemiological modelling

2020 ◽  
Author(s):  
Elba Raimúndez ◽  
Erika Dudkin ◽  
Jakob Vanhoefer ◽  
Emad Alamoudi ◽  
Simon Merkt ◽  
...  
Keyword(s):  
Monte Carlo Sampling ◽  
Critical Parameters ◽  
Parameter Estimates ◽  
Epidemiological Models ◽  
Estimation Uncertainty ◽  
Plausible Values ◽  
Calculation Algorithms ◽  
Simplifying Assumptions ◽  
Parameter Estimation Uncertainty ◽  
Substantial Uncertainty

AbstractEpidemiological models are widely used to analyse the spread of diseases such as the global COVID-19 pandemic caused by SARS-CoV-2. However, all models are based on simplifying assumptions and on sparse data. This limits the reliability of parameter estimates and predictions.In this manuscript, we demonstrate the relevance of these limitations by performing a study of the COVID-19 outbreak in Wuhan, China. We perform parameter estimation, uncertainty analysis and model selection for a range of established epidemiological models. Amongst others, we employ Markov chain Monte Carlo sampling, parameter and prediction profile calculation algorithms.Our results show that parameter estimates and predictions obtained for several established models on the basis of reported case numbers can be subject to substantial uncertainty. More importantly, estimates were often unrealistic and the confidence / credibility intervals did not cover plausible values of critical parameters obtained using different approaches. These findings suggest, amongst others, that several models are oversimplistic and that the reported case numbers provide often insufficient information.

scholarly journals Influence of parameter estimation uncertainty in Kriging: Part 1 - Theoretical Development

2001 ◽  
Vol 5 (2) ◽  
pp. 215-223 ◽  
Author(s):  
E. Todini
Keyword(s):  
Parameter Estimation ◽  
Maximum Likelihood ◽  
Covariance Matrix ◽  
Theoretical Development ◽  
Comprehensive Treatment ◽  
Parameter Estimates ◽  
Wide Field ◽  
Estimation Uncertainty ◽  
Parameter Estimation Uncertainty ◽  
Variance Covariance Matrix

Abstract. This paper deals with a theoretical approach to assessing the effects of parameter estimation uncertainty both on Kriging estimates and on their estimated error variance. Although a comprehensive treatment of parameter estimation uncertainty is covered by full Bayesian Kriging at the cost of extensive numerical integration, the proposed approach has a wide field of application, given its relative simplicity. The approach is based upon a truncated Taylor expansion approximation and, within the limits of the proposed approximation, the conventional Kriging estimates are shown to be biased for all variograms, the bias depending upon the second order derivatives with respect to the parameters times the variance-covariance matrix of the parameter estimates. A new Maximum Likelihood (ML) estimator for semi-variogram parameters in ordinary Kriging, based upon the assumption of a multi-normal distribution of the Kriging cross-validation errors, is introduced as a mean for the estimation of the parameter variance-covariance matrix. Keywords: Kriging, maximum likelihood, parameter estimation, uncertainty

scholarly journals Efficient Bayesian inference of fully stochastic epidemiological models with applications to COVID-19

2021 ◽  
Vol 8 (8) ◽  
pp. 211065
Author(s):  
Yuting I. Li ◽  
Günther Turk ◽  
Paul B. Rohrbach ◽  
Patrick Pietzonka ◽  
Julian Kappler ◽  
...  
Keyword(s):  
Bayesian Inference ◽  
Information Matrix ◽  
Monte Carlo Sampling ◽  
Epidemiological Models ◽  
Large Populations ◽  
Model Evidence ◽  
Age Structured ◽  
The Uk ◽  
Functional Central Limit

Epidemiological forecasts are beset by uncertainties about the underlying epidemiological processes, and the surveillance process through which data are acquired. We present a Bayesian inference methodology that quantifies these uncertainties, for epidemics that are modelled by (possibly) non-stationary, continuous-time, Markov population processes. The efficiency of the method derives from a functional central limit theorem approximation of the likelihood, valid for large populations. We demonstrate the methodology by analysing the early stages of the COVID-19 pandemic in the UK, based on age-structured data for the number of deaths. This includes maximum a posteriori estimates, Markov chain Monte Carlo sampling of the posterior, computation of the model evidence, and the determination of parameter sensitivities via the Fisher information matrix. Our methodology is implemented in PyRoss, an open-source platform for analysis of epidemiological compartment models.

scholarly journals Neglect Of Parameter Estimation Uncertainty Can Significantly Overestimate Structural Reliability

2015 ◽  
Vol 15 (2) ◽  
Author(s):  
Árpád Rózsás ◽  
Miroslav Sýkora
Keyword(s):  
Parameter Estimation ◽  
Bayesian Statistics ◽  
Failure Probability ◽  
Structural Reliability ◽  
Probabilistic Models ◽  
Estimation Uncertainty ◽  
Point Estimates ◽  
Distribution Parameters ◽  
Order Of Magnitude ◽  
Parameter Estimation Uncertainty

Abstract Parameter estimation uncertainty is often neglected in reliability studies, i.e. point estimates of distribution parameters are used for representative fractiles, and in probabilistic models. A numerical example examines the effect of this uncertainty on structural reliability using Bayesian statistics. The study reveals that the neglect of parameter estimation uncertainty might lead to an order of magnitude underestimation of failure probability.

On the interpretation of age–prevalence curves for trypanosome infections of tsetse flies

Parasitology ◽  
1998 ◽  
Vol 116 (2) ◽  
pp. 149-156 ◽  
Author(s):  
M. E. J. WOOLHOUSE ◽  
J. W. HARGROVE
Keyword(s):  
Experimental Data ◽  
Glossina Pallidipes ◽  
Tsetse Fly ◽  
Tsetse Flies ◽  
Parameter Estimates ◽  
Published Data ◽  
Data Sets ◽  
Trypanosoma Vivax ◽  
Epidemiological Models ◽  
Maturation Period

Epidemiological models are used to analyse 8 published data sets reporting age–prevalence curves for trypanosome infections of the tsetse fly Glossina pallidipes. A model assuming a fixed maturation period and a rate of infection which is independent of fly age is adequate for Trypanosoma vivax-type infections, explaining 98% of observed variance in prevalence by site and age, allowing that the rate of infection may be site dependent. This model is not adequate for T. congolense-type infections and the fit can be improved by allowing (i) the rates of infection to decline with age (although non-teneral flies remain susceptible), (ii) a fraction of resistant flies, which may vary between sites, (iii) increased mortality of infected flies and (iv) variation in the maturation period. Models with these features can explain up to 97% of observed variance. Parameter estimates from published experimental data suggest that all may contribute in practice but that (i) and/or (ii) are likely to be the most important.

scholarly journals When individual behaviour matters: homogeneous and network models in epidemiology

2007 ◽  
Vol 4 (16) ◽  
pp. 879-891 ◽  
Author(s):  
Shweta Bansal ◽  
Bryan T Grenfell ◽  
Lauren Ancel Meyers
Keyword(s):  
Compartmental Model ◽  
Network Models ◽  
Population Level ◽  
Disease Spread ◽  
Mixing Models ◽  
Epidemiological Models ◽  
Human Contact ◽  
Contact Patterns ◽  
Underlying Network ◽  
Simplifying Assumptions

Heterogeneity in host contact patterns profoundly shapes population-level disease dynamics. Many epidemiological models make simplifying assumptions about the patterns of disease-causing interactions among hosts. In particular, homogeneous-mixing models assume that all hosts have identical rates of disease-causing contacts. In recent years, several network-based approaches have been developed to explicitly model heterogeneity in host contact patterns. Here, we use a network perspective to quantify the extent to which real populations depart from the homogeneous-mixing assumption, in terms of both the underlying network structure and the resulting epidemiological dynamics. We find that human contact patterns are indeed more heterogeneous than assumed by homogeneous-mixing models, but are not as variable as some have speculated. We then evaluate a variety of methodologies for incorporating contact heterogeneity, including network-based models and several modifications to the simple SIR compartmental model. We conclude that the homogeneous-mixing compartmental model is appropriate when host populations are nearly homogeneous, and can be modified effectively for a few classes of non-homogeneous networks. In general, however, network models are more intuitive and accurate for predicting disease spread through heterogeneous host populations.

Sign in / Sign up

Export Citation Format

Share Document