Medium-term forecasting of socio-political stability: a mathematical model and its application to two cases

A mathematical model describing the dynamics of socio-political stability in the medium term has been developed. The main variable of the model is the level of protest activity, the empirical analogue of which is a linear combination of the number of protests and the number of repressions per unit of time. In describing the protest factors, the model takes into account the socio-economic prerequisites, and in describing the protest dynamics, the SIMCA approach is used. The model has been tested on two examples: the dynamics of protest activity in Kazakhstan and Kyrgyzstan from 2018 to 2023. It has been shown that in these cases, the model correctly describes the time of the main peaks of activity corresponding to the crisis of 2022 in Kazakhstan and the revolution of 2020 in Kyrgyzstan. The unit of time in the model is a quarter. The predictive capabilities of the model extend to the next quarter and are scenario-based, i.e. as input, the forecast requires an estimate of variables such as the inflation rate and the unemployment rate for the projected quarter.