Abstract
The devastating worldwide impact of the COVID-19 pandemic created a need to better understand the effects of vaccination on case fatality rates (CFR) in a pandemic setting. Foundational time series forecasting models (ARIMA, Prophet, LSTM) and novel hybrid models (SARIMA-Bidirectional LSTM andSARIMA-Prophet-Bidirectional LSTM) were compared for performance and accuracy to forecast vaccination inflection points for 26 countries. Correlation analyses demonstrated that stringency index, age 65 and older, life expectancy, andpositive test rate, are factors correlating the most with the vaccination and case fatality rates. The primary vaccination inflection pointwas reached at 83.27 days (15-367 days), at the vaccination rate of 13.1% (0.1% – 50%), with 42% of countries seeing the initial impact in <50 days. The secondary vaccination inflection point (SVIP) was reached at 339.31 days (161-560 days) at the cumulative vaccination rate of 67.8% (28% – 89%), with 23.1% of countries reaching it in < 300 days, 73% in the second half of 2021, and 27% in early 2022. The highest vaccination rate was achieved in Portugal (89%) and the lowest in Bulgaria (28%). All assessed machine and deep learning methodologies performed with high accuracy relative to COVID-19 historical data, demonstrated strong forecasting value, and were validated by anomaly and volatility detection analyses. The novel triple hybrid model performed the best and had the highest accuracy across all performance metrics.Countries prioritizing the health of elderly and frail population and utilizing AI technology will be better prepared for any future pandemic.
Keywords