Abstract: Intensive longitudinal data have become increasingly prevalent in studies of circadian rhythms, emotions, propagation of diseases, addictive behaviors, dyadic and family-level interactions, as well as other human dynamic processes. Such data are noisy, often multivariate in nature, and may involve multiple subjects undergoing regime switches (e.g.,showing discontinuities interspersed with continuous dynamics).

Despite increasing interest in using differential/difference equation models for representing these processes, there has been a scarcity of software packages that are fast, freely accessible, and amenable to the modeling goals of researchers of human dynamics. We have created an R package that is based on novel and computationally efficient algorithms for handling a broad class of linear and nonlinear discrete- and continuous-time models with regime-switching properties and linear Gaussian measurement functions in C, while maintaining simple and easy-to-learn model specification functions in R. We present the mathematical and computational basis used by the dynr R package, and present two illustrative examples to demonstrate the unique features of dynr.