Multivariate Weather-State Approach Conditionally Skew-Normal Generator

A multivariate weather generator for daily climate variables based on weather-states (Flecher et al. (2010) ). It uses a Markov chain for modeling the succession of weather states. Conditionally to the weather states, the multivariate variables are modeled using the family of Complete Skew-Normal distributions. Parameters are estimated on measured series. Must include the variable 'Rain' and can accept as many other variables as desired.