Evaluation applying ConvLSTM neural network for the prediction of epileptic seizures

Abstract

The COVID-19 pandemic has affected many aspects of people's lives and territories. It has a destabilizing effect on the economy due to both direct and indirect factors. In order to develop an effective policy to combat such emergencies, the problem of studying its consequences seems urgent.To form effective responses to the consequences of the pandemic, adequate scientific tools are needed to assess them. In many of the studies conducted, the impact of the pandemic was assessed without considering the previously existing socio-economic trends in the regions.The purpose of the article is to conduct a retrospective analysis of the consequences of economic instability caused by the COVID-19 pandemic on the socio-economic development of the region on the example of the Sverdlovsk region. The monthly values of twelve main socio-economic indicators from January 2016 to December 2021 were analyzed. The study is based on the use of regression models with dummy variables and seasonal autoregression for forecasting, which allow considering the seasonal dependence of socio-economic indicators.To determine the seasonality of individual indicators, a graphical method, autocorrelation correlograms and spectral analysis were used. When determining the uniformity between the predicted and actual values during the pandemic, the parametric Student's criterion and the nonparametric Wilcoxon criterion were applied, which allowed us to obtain a statistically significant result regardless of the normality of the compared aggregates.The retrospective analysis made it possible to establish a significant heterogeneous impact of the COVID-19 pandemic on most of the analyzed spheres of life in the Sverdlovsk region.