Time Series Analysis using ARIMA Models: An Approach to Forecasting Health Expenditure in USA

Nikolaos Dritsakis and Paraskevi Klazoglou
Additional contact information
Nikolaos Dritsakis: Department of Applied Informatics, University of Macedonia, Economics and Social Sciences, Thessaloniki, Greece
Paraskevi Klazoglou: Department of Applied Informatics, University of Macedonia, Economics and Social Sciences, Thessaloniki, Greece

Economia Internazionale / International Economics, 2019, vol. 72, issue 1, 77-106

Abstract: Many OECD countries are at the heart of the political agenda regarding rising healthcare spending and its long-term sustainability. The continuous rise in health expenditure exerts pressure on government budgets, health services and personal patient finance. This has led policy makers to implement reforms in order to mitigate pressures on these costs, as well as introduce programs and forecasting models to provide a support tool capable of adapting to issues that may arise. The purpose of this study is to investigate the best model to predict total health spending in the USA, a country with the highest global spending, using the Box-Jenkins methodology. Applying annual data for total US health expenditure from 1900 to 2017, resulted in the ARIMA (2,1,0) model with static forecasting being the most appropriate to predict these costs. Model estimation was achieved by the maximum likelihood-ML method and finally, the accuracy of the forecast was assessed based on certain criteria such as the root mean square error (RMSE), mean absolute percentage error (MAPE) and Theil’s inequality coefficient. Analisi di serie temporale attraverso modelli ARIMA: un approccio per la previsione della spesa sanitaria negli USA Molti paesi OCSE hanno al centro della loro agenda politica l’aumento della spesa sanitaria e la sua sostenibilità nel lungo periodo. Il continuo aumento della spesa per la sanità pubblica influenza i bilanci, i servizi sanitari così come la spesa sanitaria personale. Questo ha indotto la politica ad adottare riforme al fine di contenere questi costi e ad introdurre programmi e modelli di previsione per fornire strumenti in grado di gestire i problemi che ne possono derivare. Lo scopo di questa ricerca è trovare il modello migliore per prevedere la spesa sanitaria complessiva negli Stati Uniti, il paese con i costi maggiori, utilizzando la metodologia Box-Jenkins. Applicando dati annuali relativi alla spesa sanitaria totale degli USA dal 1900 al 2017, il modello a previsione statica più appropriato è risultato essere ARIMA (2,1,0). La stima del modello è stata effettuata col metodo di massima verosimiglianza e l’accuratezza della previsione è stata valutata tramite l’errore a radice quadratica (RMSE), l’errore a percentuale media assoluta (MAPE) e il coefficiente di disuguaglianza di Theil.

Keywords: ARIMA Model; Health Expenditure; Box-Jenkins; Forecasting (search for similar items in EconPapers)
JEL-codes: C53 E27 (search for similar items in EconPapers)
Date: 2019
References: View references in EconPapers View complete reference list from CitEc
Citations: View citations in EconPapers (3)

Downloads: (external link)
http://www.iei1946.it/upload/rivista_articoli/alle ... itsakisricfinalx.pdf Full text (application/pdf)

Related works:
This item may be available elsewhere in EconPapers: Search for items with the same title.

Export reference: BibTeX RIS (EndNote, ProCite, RefMan) HTML/Text

Persistent link: https://EconPapers.repec.org/RePEc:ris:ecoint:0841

Access Statistics for this article

Economia Internazionale / International Economics is currently edited by Giovanni Battista Pittaluga

More articles in Economia Internazionale / International Economics from Camera di Commercio Industria Artigianato Agricoltura di Genova Via Garibaldi 4, 16124 Genova, Italy. Contact information at EDIRC.
Bibliographic data for series maintained by Angela Procopio ().