Open Access Article SciPap-814
Comparison of Bayesian Estimates of Health Care Costs Fot Czech and Sovakian Insurance Companies
by Pavla Jindrová 1,* and Kateřina Seinerová 2

1 Faculty of Economics and Administration, Institute of Mathematics and Quantitative Methods, University of Pardubice, Studenská 84, Pardubice 53210, Czechia

2 Faculty of Economics and Administration, Institute of Mathematics and Quantitative Methods, University of Pardubice, Studenská 84, Pardubice 53210, Czechia

* Authors to whom correspondence should be addressed.

Abstract: This paper presents an application of the Bayesian inference to estimate health care expenses by health insurance companies in the Czech Republic and Slovakia. In particular, the Buhlmann-Straub model is applied onto real data, covering time period 2010-2015, from all seven Czech and three Slovakian health insurance companies in order to calculate the credible health care costs per person and the corresponding credibility factors for each company. Computed credible health care costs per person are compared with the real estimations of health care costs per person quoted by analyzed health insurance companies. These real estimates are not available for all the Slovakian health insurance companies and two Czech health insurance companies. The results of Buhlmann-Straub model show that while the largest insurance companies can reliably predict the health care costs for the subsequent time period based on its own statistical data, the smaller companies can find out credible estimation of health care costs per person, taking into account data from all comparable healthcare insurance companies.

Keywords: Health Insurance, Bayesian Estimation, Credibility Model, Bühlmann-Straub Model, Health Insurance Companies

JEL classification:   C11 - Bayesian Analysis: General,   C22 - Time-Series Models • Dynamic Quantile Regressions • Dynamic Treatment Effect Models • Diffusion Processes,   G22 - Insurance • Insurance Companies • Actuarial Studies

SciPap 2016, 24(2), 814

Received: 1 May 2015 / Accepted: 8 September 2016 / Published: 16 September 2016