?url_ver=Z39.88-2004&rft_val_fmt=info%3Aofi%2Ffmt%3Akev%3Amtx%3Adc&rft.relation=http%3A%2F%2Feprints.imtlucca.it%2F4076%2F&rft.title=Simulation+of+Covid-19+epidemic+evolution%3A%0D%0Aare+compartmental+models+really+predictive%3F&rft.creator=Paggi%2C+Marco&rft.subject=HA+Statistics&rft.subject=TJ+Mechanical+engineering+and+machinery&rft.description=Computational+models+for+the+simulation+ofthe+severe+acute+respiratory+syndrome%0D%0Acoronavirus+2+(SARS-CoV-2)+epidemic+evolution+would+be+extremely+useful+to%0D%0Asupport+authorities+in+designing+healthcare+policies+and+lockdown+measures+to%0D%0Acontain+its+impact+on+public+health+and+economy.+In+Italy%2C+the+devised+forecasts%0D%0Ahave+been+mostly+based+on+a+pure+data-driven+approach%2C+by+fitting+and%0D%0Aextrapolating+open+data+on+the+epidemic+evolution+collected+by+the+Italian+Civil%0D%0AProtection+Center.+In+this+respect%2C+SIR+epidemiological+models%2C+which+start+from+the%0D%0Adescription+of+the+nonlinear+interactions+between+population+compartments%2C%0D%0Awould+be+a+much+more+desirable+approach+to+understand+and+predict+the+collective%0D%0Aemergent+response.+The+present+contribution+addresses+the+fundamental+question%0D%0Awhether+a+SIR+epidemiological+model%2C+suitably+enriched+with+asymptomatic+and%0D%0Adead+individual+compartments%2C+could+be+able+to+provide+reliable+predictions+on+the%0D%0Aepidemic+evolution.+To+this+aim%2C+a+machine+learning+approach+based+on+particle%0D%0Aswarm+optimization+(PSO)+is+proposed+to+automatically+identify+the+model%0D%0Aparameters+based+on+a+training+set+of+data+of+progressive+increasing+size%2C%0D%0Aconsidering+Lombardy+in+Italy+as+a+case+study.+The+analysis+of+the+scatter+in+the%0D%0Aforecasts+shows+that+model+predictions+are+quite+sensitive+to+the+size+of+the+dataset%0D%0Aused+for+training%2C+and+that+further+data+are+still+required+to+achieve+convergent+-%0D%0Aand+therefore+reliable-+predictions.&rft.publisher=IMT+School+for+Advanced+Studies+Lucca&rft.date=2020-04-16&rft.type=Working+Paper&rft.type=NonPeerReviewed&rft.format=application%2Fpdf&rft.language=en&rft.rights=cc_by&rft.identifier=http%3A%2F%2Feprints.imtlucca.it%2F4076%2F1%2FTR_11_2020_Paggi.pdf&rft.identifier=++Paggi%2C+Marco++Simulation+of+Covid-19+epidemic+evolution%3A+are+compartmental+models+really+predictive%3F++CSA+Technical+Report++%2311%2F2020++++IMT+School+for+Advanced+Studies+Lucca+%2C+Lucca.++++++++