High volatility, high emissions? a hidden-Markov model approach

This study introduces the generalised normal distribution hidden Markov model (GND-HMM) with constrained and unconstrained parameters, where a first-order Markov chain governs the draws of the states from the mixture components. The proposed model is applied to the daily electricity price returns of the electricity market in northern Italy to detect the tur- moil periods that occurred during the years 2020-2023. The turmoil periods detected by the GND-HMM model highlight important aggregate events such as the Covid-19 pandemic and the Russia-Ukranian conflict. Furthermore, our study aims to examine the relationship between the identified turmoil periods and the time series of CO2 emissions in the northern Italian electricity market.