seabbs/SpeedyMarkov

Speed up Discrete Markov Model Cost Effectiveness Simulations. This package: Compares a functional markov modelling approach to a reference approach for several example models; Explores approaches to speeding up Markov modelling in a principled fashion making use of C++ when required; Details the benefits of parallisation and provide a code structure in which parallisation is easy to make use of; Provides a toolkit for use in discrete Markov modelling. Provides optimised code that may be ported into other applications and workflows. The work in this package was started at the Health Economic 2019 hackathon hosted at Imperial. Much of this work is based on that developed by the hermes6 team. The original reference approach was developed by Howard Thom.

Keywords

Discrete Markov Model