A Dynamic Model Averaging Approach -COVID-19

 

Forecasting the Spread of the COVID-19 Epidemic in Lombardy: A Dynamic Model Averaging Approach

Lisa Gianmoena, View ORCID ProfileVicente Rios

doi: https://doi.org/10.1101/2021.01.18.21250053

This article is a preprint and has not been peer-reviewed [what does this mean?]. It reports new medical research that has yet to be evaluated and so should not be used to guide clinical practice.

 

Forecasting with accuracy the evolution of COVID-19 daily incidence curves is one of the most important exercises in the field of epidemic modeling. We examine the forecastability of daily COVID-19 cases in the Italian region of Lombardy using Dynamic Model Averaging and Dynamic Model Selection methods. To investigate the predictive accuracy of this approach, we compute forecast performance metrics of sequential out-of-sample real-time forecasts in a back-testing exercise ranging from March 1 to December 10 of 2020. We find that (i) Dynamic Model Averaging leads to a consistent and substantial predictive improvements over alternative epidemiological models and machine learning approaches when producing short-run forecasts. Using estimated posterior inclusion probabilities we also provide evidence on which set of predictors are relevant for forecasting in each period. Our findings also suggest that (ii) future incidences can be forecasted by exploiting information on the epidemic dynamics of neighboring regions, human mobility patterns, pollution and temperatures levels.

Competing Interest Statement

The authors have declared no competing interest.

Funding Statement

This work has been financially supported by the Bando Dipartimenti di Eccellenza Grant, awarded by the Italian ministry of education (MIUR) to the Centre of Excellence in Economics and Data Science (CEEDS)of the University of Milan.

Author Declarations

I confirm all relevant ethical guidelines have been followed, and any necessary IRB and/or ethics committee approvals have been obtained.

Yes

The details of the IRB/oversight body that provided approval or exemption for the research described are given below:

No IRB/oversight body has provided approval for the research.

All necessary patient/participant consent has been obtained and the appropriate institutional forms have been archived.

Yes

I understand that all clinical trials and any other prospective interventional studies must be registered with an ICMJE-approved registry, such as ClinicalTrials.gov. I confirm that any such study reported in the manuscript has been registered and the trial registration ID is provided (note: if posting a prospective study registered retrospectively, please provide a statement in the trial ID field explaining why the study was not registered in advance).

Yes

I have followed all appropriate research reporting guidelines and uploaded the relevant EQUATOR Network research reporting checklist(s) and other pertinent material as supplementary files, if applicable.

Yes

Paper in collection COVID-19 SARS-CoV-2 preprints from medRxiv and bioRxiv

Copyright 

The copyright holder for this preprint is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. It is made available under a CC-BY-NC-ND 4.0 International license.

-

 

Segundo paper de Lisa Gianmoena y Vicente Rios de la saga Covid19 mejorando la calidad de los pronósticos a corto plazo

-

Pronosticar con precisión la evolución de las curvas de incidencia diaria COVID-19 es uno de los ejercicios más importantes en el campo del modelado epidemico. Examinamos la previsión de los casos diarios de COVID-19 en la región italiana de Lombardía utilizando métodos dinámicos de media y modelos dinámicos de selección de modelos. Para investigar la precisión predecible de este enfoque, calificamos las medicinas de rendimiento de las previsiones secuenciales fuera de la muestra en tiempo real en un ejercicio de prueba que va desde el 1 de marzo hasta el 10 de diciembre de 2020. Encontramos que (i) El promedio de modelos dinámicos lleva a una mejora predictiva consistente y sustancial de los modelos epidemiológicos alternativos y los enfoques de aprendizaje automático cuando se producen pronósticos a corto plazo. Usando las probabilidades de inclusión posterior estimadas también proporcionamos pruebas sobre qué conjunto de predicadores son relevantes para la previsión en cada período. Nuestras conclusiones también sugieren que ii) las futuras incidencias pueden predecirse mediante la explotación de la información sobre la dinámica epidemia de las regiones vecinas, los patrones de movilidad humana, la contaminación y los niveles de temperatura.