DOI: 10.11118/ecos2026_003

ECONOMETRIC MODEL OF GENERALIZED LINEAR REGRESSION FOR PREDICTING FUTURE NUMBERS OF INSURED PEOPLE BY GENERAL HEALTH INSURANCE COMPANY OF THE CZECH REPUBLIC – COMPARISON OF RESULTS AFTER FIVE YEARS

Jaromír Běláček¹

¹ Department of Statistics and Operation Analysis, Faculty of Business and Economics, Mendel University in Brno, Zemědělská 1, 613 00 Brno, Czech Republic

This paper compares the results of predicting the number of insured persons of the General Health Insurance Company of the Czech Republic in 2018, based on a generalized-discounted linear regression model (GLM), which formalizes the projected number of insured persons in age groups through exogenous demographic, re-reregistration, and formal linear and quadratic trend components, with reality observed in 2018–2022. Compared to standard in-sample methods, derived population projections, or the traditional indexing method, the 2018 method, where the calibration of discount factors α meets the criteria for maximizing the determination coefficient R2(α), overall yields better results. Accurate predictions of future numbers of insured individuals have the potential to better calculate and plan future expenditures of the health insurance funds during current period of population aging, and for several years ahead, too.

Klíčová slova: public health insurance, time series forecasting, derived population projection, exponential smoothing, coefficient of determination, weighted-discounted least squares

stránky: 4-4, online: 2026