Robust Determinants of Health Sector Costs in Iran: Bayesian Model Averaging Approach

Subject Areas : medical documents

moohamad Alizadeh ¹ , abolghasem Golkhandan ²

1 - Assistant Professor Department of Economics, Faculty of Economics and Administrative, Lorestan University, Khoram Abad, Iran
2 - Ph.D student Economics, Faculty of Economics and Administrative, Lorestan University, Khoram Abad, Iran

Received: 2016-06-19 Accepted : 2016-12-20 Published : 2016-08-22

Keywords: Iran, Health Sector Costs, Bayesian Model Averaging Approach, Uncertainty of Model,

Abstract :

Introduction: Identify of factors that influence on health costs can be useful in determine the best policy to control and manage the health costs. Previous studies in this area has been done with assumption the certainty of model; While the lack of attention to the problem of model uncertainty can lead to bias and lack of performance in estimation of parameters that result is inappropriate forecasts and incorrect statistical inference. So, the main objective of this study is identify the robust determinants of health sector costs in Iran under uncertainty of model. Methods: This study uses the statistical data of 22 variables that affect health sector costs based on theoretical and empirical studies, is paid to identify the robust determinants of these costs in Iran during 1979-2013. For this purpose is used the Bayesian Averaging of Classical Estimates (BACE) approach (due to favorable characteristics for the assumption of model uncertainty). Also, the statistical analyzes were performed using the R software. Results: estimation of 40000 regression and Bayesian averaging from the coefficients shows that per capita income with the possibility of 0.98 and coefficient of 0.70, urbanization rate with the possibility of 0.93 and coefficient of 1.25, per capita public health costs with the possibility of 0.83 and coefficient of 0.29, dependency ratio with the possibility of 0.50 and coefficient of 0.27, physician per capita with the possibility of 0.49 and coefficient of 0.20 and the unemployment rate with the possibility of 0.38 and coefficient of -0.07, are non-fragile and robust variables. Conclusion: The results indicate that the most important determinants of health sector costs in Iran are respectively: per capita income, urbanization rate, per capita public health costs, dependency ratio, physician per capita and unemployment rate. The effect of all these variables on per capita health sector costs in the long run are sure and strong.  

References:

1- Newhouse J. Medical care expenditures; a cross national study, J Hum Resource, 1977; 12: 10-26.

2- Magazzino C and Mele M. The Determinants of health expenditure in Italian regions, International Journal of Economics & Finance, 2012; 4(3): 61-72. 

3- Wang Z. The determinants of health expenditures: evidence from US state-level data, Applied Economics, 2009; 41(4): 429-35.

4- Ang JB. The determinants of health care expenditure in Australia, Applied Economics Letters, 2009; 17(4): 639-44.

5- Bilgel F and Tran. K.C. The determinants of Canadian provincial health expenditures: evidence from a dynamic panel, 2012; 45(2): 201-12.

6- Hosoya, K. Determinants of health expenditures: Stylized facts and a new signal, Modern Economy, 2014; 5: 1171-80.

7- Fattahi M, Osari A, Sadegi H and Asgharpur H. Effects of air pollution on public spending for health: Comparative developing and developed countries, Journal of Economic Development, 2013; 3(11): 111-32. [In Persian]

8- Rezaei S, Dindar A, Rezapour A. health care expenditures and their determinants: Iran provinces (2006-2011), Journal of Health Administration, 2016; 19 (63): 81-90. [In Persian]

9- Liu, C and Maheu, J. M. Forecasting realized volatility: A Bayesian Model-Averaging approach, Article first published online: Journal of Applied Econometrics, 2009; 22: 4-6.

10- Draper, D. Assessment and propagation of model uncertainty, Journal of the Royal Statistical Society, 1995; 57: 45-70.

11- Sala-i-Martin X, Doppelhofer G and Miller R. Determinants of long-Term growth: A Bayesian Averaging of Classical Estimates (BACE) approach, The American economic review, 2004; 94: 813-35.

12- Leamer E. Let's take the con Out of econometrics, American Economic Review, 1983; 73: 31-43.

13- Noble RB. Multivariate applications of Bayesian Model Averaging, Working Paper, 2000.

14- George EI, McCulloch RE. Variable selection via Gibbs sampling, Journal of the American Statistical Association, 1993; 88: 881-889.