Comparison of ARIMA, ETS, NNAR, TBATS and hybrid models to forecast the second wave of COVID-19 hospitalizations in Italy.

The coronavirus disease (COVID-19) is a severe, ongoing, novel pandemic that emerged in Wuhan, China, in December 2019. As of January 21, 2021, the virus had infected approximately 100 million people, causing over 2 million deaths. This article analyzed several time series forecasting methods to predict the spread of COVID-19 during the pandemic's second wave in Italy (the period after October 13, 2020). The autoregressive moving average (ARIMA) model, innovations state space models for exponential smoothing (ETS), the neural network autoregression (NNAR) model, the trigonometric exponential smoothing state space model with Box-Cox transformation, ARMA errors, and trend and seasonal components (TBATS), and all of their feasible hybrid combinations were employed to forecast the number of patients hospitalized with mild symptoms and the number of patients hospitalized in the intensive care units (ICU). The data for the period February 21, 2020-October 13, 2020 were extracted from the website of the Italian Ministry of Health ( www.salute.gov.it ). The results showed that (i) hybrid models were better at capturing the linear, nonlinear, and seasonal pandemic patterns, significantly outperforming the respective single models for both time series, and (ii) the numbers of COVID-19-related hospitalizations of patients with mild symptoms and in the ICU were projected to increase rapidly from October 2020 to mid-November 2020. According to the estimations, the necessary ordinary and intensive care beds were expected to double in 10 days and to triple in approximately 20 days. These predictions were consistent with the observed trend, demonstrating that hybrid models may facilitate public health authorities' decision-making, especially in the short-term.

Assessing the impact of long-term exposure to nine outdoor air pollutants on COVID-19 spatial spread and related mortality in 107 Italian provinces.

This paper investigates the air quality in 107 Italian provinces in the period 2014-2019 and the association between exposure to nine outdoor air pollutants and the COVID-19 spread and related mortality in the same areas. The methods used were negative binomial (NB) regression, ordinary least squares (OLS) model, and spatial autoregressive (SAR) model. The results showed that (i) common air pollutants-nitrogen dioxide (NO2), ozone (O3), and particulate matter (PM2.5 and PM10)-were highly and positively correlated with large firms, energy and gas consumption, public transports, and livestock sector; (ii) long-term exposure to NO2, PM2.5, PM10, benzene, benzo[a]pyrene (BaP), and cadmium (Cd) was positively and significantly correlated with the spread of COVID-19; and (iii) long-term exposure to NO2, O3, PM2.5, PM10, and arsenic (As) was positively and significantly correlated with COVID-19 related mortality. Specifically, particulate matter and Cd showed the most adverse effect on COVID-19 prevalence; while particulate matter and As showed the largest dangerous impact on excess mortality rate. The results were confirmed even after controlling for eighteen covariates and spatial effects. This outcome seems of interest because benzene, BaP, and heavy metals (As and Cd) have not been considered at all in recent literature. It also suggests the need for a national strategy to drive down air pollutant concentrations to cope better with potential future pandemics.

Using the SARIMA Model to Forecast the Fourth Global Wave of Cumulative Deaths from COVID-19: Evidence from 12 Hard-Hit Big Countries

The COVID-19 pandemic is a serious threat to all of us. It has caused an unprecedented shock to the world's economy, and it has interrupted the lives and livelihood of millions of people. In the last two years, a large body of literature has attempted to forecast the main dimensions of the COVID-19 outbreak using a wide set of models. In this paper, I forecast the short- to mid-term cumulative deaths from COVID-19 in 12 hard-hit big countries around the world as of 20 August 2021. The data used in the analysis were extracted from the Our World in Data COVID-19 dataset. Both non-seasonal and seasonal autoregressive integrated moving averages (ARIMA and SARIMA) were estimated. The analysis showed that: (i) ARIMA/SARIMA forecasts were sufficiently accurate in both the training and test set by always outperforming the simple alternative forecasting techniques chosen as benchmarks (Mean, Naïve, and Seasonal Naïve);(ii) SARIMA models outperformed ARIMA models in 46 out 48 metrics (in forecasting future values), i.e., on 95.8% of all the considered forecast accuracy measures (mean absolute error [MAE], mean absolute percentage error [MAPE], mean absolute scaled error [MASE], and the root mean squared error [RMSE]), suggesting a clear seasonal pattern in the data;and (iii) the forecasted values from SARIMA models fitted very well the observed (real-time) data for the period 21 August 2021–19 September 2021 for almost all the countries analyzed. This article shows that SARIMA can be safely used for both the short- and medium-term predictions of COVID-19 deaths. Thus, this approach can help government authorities to monitor and manage the huge pressure that COVID-19 is exerting on national healthcare systems.

The determinants of COVID-19 case fatality rate (CFR) in the Italian regions and provinces: An analysis of environmental, demographic, and healthcare factors.

The Italian government has been one of the most responsive to COVID-2019 emergency, through the adoption of quick and increasingly stringent measures to contain the outbreak. Despite this, Italy has suffered a huge human and social cost, especially in Lombardy. The aim of this paper is dual: i) first, to investigate the reasons of the case fatality rate (CFR) differences across Italian 20 regions and 107 provinces, using a multivariate OLS regression approach; and ii) second, to build a "taxonomy" of provinces with similar mortality risk of COVID-19, by using the Ward's hierarchical agglomerative clustering method. I considered health system metrics, environmental pollution, climatic conditions, demographic variables, and three ad hoc indexes that represent the health system saturation. The results showed that overall health care efficiency, physician density, and average temperature helped to reduce the CFR. By the contrary, population aged 70 and above, car and firm density, air pollutants concentrations (NO2, O3, PM10, and PM2.5), relative average humidity, COVID-19 prevalence, and all three indexes of health system saturation were positively associated with the CFR. Population density, social vertical integration, and altitude were not statistically significant. In particular, the risk of dying increases with age, as 90 years old and above had a three-fold greater risk than the 80-to-89 years old and four-fold greater risk than 70-to-79 years old. Moreover, the cluster analysis showed that the highest mortality risk was concentrated in the north of the country, while the lowest risk was associated with southern provinces. Finally, since prevalence and health system saturation indexes played the most important role in explaining the CFR variability, a significant part of the latter may have been caused by the massive stress of the Italian health system.