Skip to main navigationSkip to main content
The University of Southampton
CORMSIS Centre for Operational Research, Management Sciences and Information Systems

CORMSIS Seminar "Modelling intermittent time series and forecasting Covid-19 spread in the USA" - Giacomo Sbrana Event

Date:
4 March 2021
Venue:
Please email Huan Yu for a link to the virtual seminar

For more information regarding this event, please email Huan Yu at Huan.Yu@southampton.ac.uk .

Event details

Forecasting intermittent time series represents a challenging task whose importance grows together with the increasing number of series observed sporadically. However, given the difficulties in modeling the presence of zeros, few methods are available. This paper introduces a novel state-space model defined as Intermittent Exponential Smoothing (IES). Our approach allows encompassing the intermittent nature of time series and forecasting efficiently. Indeed, the proposed state-space approach integrates the random nature of the series assuming a Bernoulli dynamics that allows switching between zeros and positive values. Moreover, we derive the unobserved dynamics of the time series and provide a simple method for estimating and forecasting future sporadic observations. Finally, we compare the performance of our method with those of standard intermittent models using the daily number of new cases of Covid-19 observed in more than 3000 American counties. Predicting the number of newly infected people is extremely important, not only for hospitals but also for policy makers in general. Empirical results show that the suggested approach clearly outperforms all its competitors in forecasting the number of new Coronavirus cases over a period of ten days.

Speaker information

Dr. Sbrana Giacomo, Full Professor at NEOMA Business School since 2019. He joined the school in 2011 as Assistant Professor and became Associate Professor in 2014. Before joining the school, he was post-doc at BETA-CNRS, University of Strasbourg (France) and consultant for the OECD, Paris. Previously he was economist at the Department of Economic and Social Affairs of the United Nations (New York, 2005-2009). He holds a Master of Science in Economics and Econometrics from the University of Southampton (U.K.) and a Ph.D in Statistics from the University of Roma Tre (Italy). Professor Sbrana primary research interests include time series analysis and forecasting especially using State-Space models and the Kalman filter. He has published in several academic journals such as: Journal of Time Series Analysis, Journal of Forecasting, Journal of Banking & Finance, Economic Modelling, International Journal of Production Economics, International Journal of Forecasting, Journal of Multivariate Analysis, Macroeconomic Dynamics, Cliometrica, Bulletin of Economic Research, Journal of the Operational Research Society

Privacy Settings