The small library of recursive methods in R (librec) is a set of functions to solve sequential problems that can be represented by first-order stochastic linear difference equations and Bellman functional equations, namely, the optimal linear quadratic regulator problem (OLRP) or the Nash-Markov linear quadratic game. This library is an adaptation to R of the MatLab toolkit developed by Thomas J. Sargent, the recipient of the 2011 Nobel Prize in Economic Sciences, in the scope of the textbooks Recursive Models of Dynamic Linear Economies (Hansen and Sargent, 2013) and Recursive Macroeconomic Theory (Ljungqvist and Sargent, 2018).
In addition, librec is enriched with routines to filter data (e.g. Hodrick-Prescott and Mosheiov and Raveh trend-cycle decomposition) and reinforcement learning (RL) time differences (TD) algorithms.
A playground with examples will be also provided – soon available.