Enhancing Inferential Accuracy with Bootstrap Methods: A Statistical Approach to Insurance Data in Urban Planning Contexts

Sufian Munther  Salih

Sufian Munther Salih Department of Management Economics Banking, College of Bussiness Economics, AlNahrain University, Baghdad, Iraq

Keywords: : bootstrap method, inferential statistics, regression estimation, regional planning, data-driven planning

Abstract

In applied statistics, particularly in domains like economics and insurance, small sample sizes and non-ideal data conditions often compromise the accuracy of traditional inferential methods. Iraqi insurance sector data from 1999 to 2014 offers only 16 observations, making classical regression approaches unsuitable due to their dependence on large sample assumptions. There is insufficient understanding of how bootstrapping methods compare in terms of estimation reliability under such constrained data conditions. This study aims to assess the effectiveness of bootstrap resampling both error-based and observation-based in estimating regression parameters related to premium retention rates in the Iraqi insurance industry. Empirical comparisons using mean squared error (MSE), mean absolute error (MAE), and mean absolute percentage error (MAPE) reveal that the error resampling method significantly outperforms the observation resampling method in fitting accuracy. The study further identifies six key predictors of premium retention rate, including corporate capital, changes in underwriting, population size, bank credit, bank deposits, and education levels (risk aversion). This research uniquely applies bootstrap methods to an underexplored dataset within the insurance sector of a developing country, demonstrating how inferential robustness can be achieved without reliance on large samples or strict distributional assumptions. The findings support the broader adoption of error-based bootstrap techniques in policy modeling and financial forecasting, particularly in data-scarce environments common to developing economies and urban planning contexts.

Downloads

Download data is not yet available.

References

M. H. Ahmed, "Towards a quantitative model for determining the optimal retention limit and its impact on the probability of company collapse," New Horizons Journal, Faculty of Commerce, Menoufia University, vol. 10, no. 1, 1998.

M. M. El-Derini, "Using the statistical bootstrap method in selecting the optimal model for the private savings function in Egypt," J. Egyptian Stat. Soc. (JESS), vol. 21, no. 1, 2005.

B. M. El-Omari, Treating the problem of heterogeneity of variances using the statistical bootstrap method, M.S. thesis, Faculty of Commerce, Tanta University, 2001.

A. H. A. El-Qadhi, "A study of the absorptive capacity of the insurance market," Academy of Scientific Research, Risk and Insurance Research Division, 2007.

M. A. R. Ismail, Linear Regression Analysis. Riyadh: Institute of Public Administration, Research Center, 2001.

H. Hussein and N. El-Hamidi, "Using the retention rate and the utilized capacity ratio in insurance portfolio risk analysis," presented at the 7th Int. Forum on the Insurance Industry, Hassiba University, Faculty of Economics, Business and Management Sciences, 2012.

M. Abdul Mawla and M. Al-Mahdi, "Towards an integrated framework for determining the factors affecting retention limits in life insurance: An applied study on Arab direct insurance companies," New Horizons J. Bus. Stud., Faculty of Commerce, Menoufia University, no. 2, 1995.

M. A. G. Muawad, A proposed statistical model for predicting Egyptian cotton exports, Ph.D. dissertation, Faculty of Commerce, Ain Shams University, 2007.

S. S. D. El-Din, "Using the bootstrap method to find new estimators for the ratio in simple random sampling," Iraqi J. Statist. Sci., vol. 16, 2009.

B. Efron, "Bootstrap methods: Another look at the jackknife," Ann. Statist., vol. 7, no. 1, pp. 1–26, 1979.

D. A. Freedman, "Bootstrapping regression models," Ann. Statist., vol. 9, no. 6, pp. 1218–1228, 1981.

J. Fox, Applied Regression Analysis, Linear Models and Related Methods. Thousand Oaks, CA: Sage Publications, 1997.

Y. R. Liu, "Bootstrap procedures under some non-i.i.d. models," Ann. Statist., 1988.

J. Shao, "Bootstrap model selection," J. Am. Stat. Assoc., vol. 91, no. 434, pp. 655–665, Jun. 1996.

J. Shao and D. Tu, The Jackknife and Bootstrap. New York: Springer, 1995.

S. Sahinler and D. Topuz, "Bootstrap and jackknife resampling algorithms for estimation parameters," J. Appl. Quant. Methods, vol. 2, no. 2, pp. 187–197, 2007.

R. Stine, "Bootstrap prediction intervals for regression," J. Am. Stat. Assoc., vol. 80, pp. 1026–1031, 1985.

R. Stine, "Modern methods of data analysis," in Modern Methods of Data Analysis, J. Fox, Ed. Scotland: Sage Publications, 1990, pp. 325–373.