One of the key research products of Project AERO is the R package spaero (pronounced “sparrow”) for estimating distributional properties along rolling windows of time series. Such estimates may in some cases provide signals that the system generating the data is approaching a regime shift. Regime shifts are marked changes in the dynamics of a system that may sometimes be modeled as bifurcations of dynamical systems. Examples of such regime shifts in nature include the eutrophication of lakes, changes in climate, and the emergence or eradication of infectious diseases. Because these methods are based on generic properties of dynamical systems, they have the potential to apply to a broad range of models. The spaero package also provides functions to simulate epidemics and support computational experiments designed to evaluate these methods for application to infectious disease systems.
Version 0.3 is now available on the CRAN repository at https://cran.r-project.org/web/packages/spaero/. Source code, binaries, and introductory documentation (i.e., vignette) are available.