Code

MSTest - R package
with Jean-Marie Dufour

This package implements hypothesis testing procedures that can be used to identify the number of regimes in a Markov-Switching model. It includes the methods proposed in Hansen (1992), Carrasco et al. (2014), Dufour & Luger (2017), and Rodriguez-Rondon & Dufour (2024).

mbreaks v2.0 - R package
with Linh Nguyen, Pierre Perron, & Yohei Yamamoto

This package implements the efficient dynamic programming approach to conduct estimation and testing for linear models in presence of structural breaks in the mean and variance as described in Bai & Perron (1998) and Perron, Yamamoto, & Zhou (2020).

bootBtest - R package

This package allows the user to perform the bootstrap pre-test procedure described in Davidson & MacKinnon (2000) to determine an appropriate number of bootstrap simulations to use to minimize loss of power resulting from sampling.

MaxMC - R package
with Julien Neves & Jean-Marie Dufour

An implementation of the Monte Carlo techniques described in details by Dufour (2006) and Dufour and Khalaf (2007). The two main features available are the Monte Carlo method with tie-breaker, mc(), for discrete statistics, and the Maximized Monte Carlo, mmc(), for statistics with nuisance parameters.