Relations between statistically and mechanistically derived models: Analysis of a food-chain with complicated dynamics
Year: 1999
Type: report
Abstract: In the qualitative part of this paper we present substantial evidence for that the dynamical patterns of food-chains predicted by Gragnani, DeFeo and Rinaldi (Bull.Math.Biol., 1998) are robust with respect to discrete birth processes.Our theoretical study contains a rigorous proof of the discrete Hopf-bifurcation of the interior fixed point of the three dimensional system using the Schur-Cohn and the Routh-Hurwitz criteria.A detailed numerical investigation establishes the predicted patterns for non-equilibrium cases.The statistical part is motivated by the increasing interest for using statistical methods outside their basic scope.Examples may include predicting number of species taking part in a given ecology or Lyapunov exponents from biological time-series.We fit both autoregressive models and multivariate adaptive regression splines (MARS) to data generated by our theoretical model and compare the statistical properties of the time series to our theoretical analysis.
Author: Torstein Lindström
Language: English
Subfield: Applied Mathematics
Field: Mathematics
Domain: Physical Sciences
Open Access status: green
Funders: European Commission, Norges Forskningsråd
OpenAlex ID: W2613093740
