Related Experiment Videos
Interpreting time-series analyses for continuous-time biological models--measles as a case study.
1Department of Zoology, University of Cambridge, Downing Street, Cambridge CB2 3EJ, UK. katie@zoo.cam.ac.uk
Journal of Theoretical Biology
|June 5, 2003
Summary
Discretizing continuous-time ecological models for measles requires a non-homogeneous contact function. This adjustment is crucial for accurately modeling host-parasite interactions in time-series data, even without spatial or age structure.
Area of Science:
- Ecological modeling
- Epidemiology
- Mathematical biology
Background:
- Mechanistic models are increasingly fitted to ecological time-series data.
- Discrete-time models are often used as approximations of continuous-time processes.
- Measles time-series models are being developed to estimate ecological parameters.
Purpose of the Study:
- To test the validity of discretizing continuous-time mechanistic models for ecological time-series.
- To investigate the specific case of measles dynamics.
- To identify conditions under which discretization may fail.
Main Methods:
- Fitting discrete-time mechanistic models to measles time-series data.
- Developing a non-homogeneous contact function to capture host-parasite interactions.
- Deriving a mathematical relationship for the departure from mass-action transmission.
Main Results:
- A non-homogeneous contact function is essential for discrete-time measles models, even without spatial or age structure.
- A mathematical relationship was derived to quantify deviations from mass-action transmission.
- Conditions were identified where the discretization process may be invalid.
Conclusions:
- Discretization of continuous-time models for ecological time-series, particularly for measles, requires careful consideration of the contact function.
- The assumption of mass-action transmission may not hold in discrete-time approximations.
- Accurate ecological parameter estimation from time-series data depends on appropriate model formulation.