About the Project
Summary
Researchers collect data on health outcomes in order to assess variations in short term and long term health status of infants and adults in developing and developed countries. Key statistical hypotheses regarding geographical variation and intertemporal variation in health inequalities are typically evaluated using large sample methods of statistical inference. These methods may not be accurate in finite samples, potentially leading the researcher to reject true hypotheses more often than the researcher’s choice of a significance level suggests.
Throughout this research, we shall evaluate the accuracy of alternative methods of statistical inference using Monte Carlo experiments. For each given method of inference, we shall examine empirical rejection probabilities as well as statistical power. Firstly, we examine each given method of inference by comparing the empirical rejection probability with a nominal rejection probability (i.e. significance level). For each method of interest, we shall repeatedly simulate random samples from a population, and we shall evaluate the frequency with which a known true hypothesis is incorrectly rejected at a given significance level. Likewise, our second concern will be to evaluate statistical power: the frequency with which a false hypothesis is rejected.
The difference between the empirical and nominal rejection probabilities is known as the error rejection probability (ERP). The larger the ERP in magnitude, the less accurate a method of inference is deemed in finite samples. Likewise, the lower the statistical power, the less desirable a statistical method of inference is deemed.
Objectives
The first objective of this research is to evaluate the ERP and statistical power of asymptotic inference methods as well as simulation based methods in the context of inequality measures of self-assessed health.
The second objective of this research is to evaluate the ERP and statistical power of asymptotic inference methods as well as simulation based methods in the context of inequality measures of socioeconomic inequalities in health.
The third objective of this project is to apply, within the context of a demographic and health survey, the different methods of inference and to illustrate how they may produce contrasted results regarding the temporal and geographical variation in health inequalities.
Prerequisites
A strong background in econometrics and statistics, at the level of Davison A and D. Hinkley (1997): Bootstrap methods and their application, Cambridge University Press.
Funding Notes
This project is funded by a University of Aberdeen Elphinstone Scholarship. An Elphinstone Scholarship covers the cost of tuition fees, whether Home, EU or Overseas.
Selection will be made on the basis of academic merit.
References
Abul Naga R. and C. Stapenhurst (2015): “Estimation of inequality indices of the cumulative distribution function”, Economics Letters 130, 109-112.
Abul Naga R. Shen and H-I Yoo (2016): “Joint hypotheses tests for multidimensional inequality indices”, Economics Letters 141, 138-142.
Biewen, M. (2002): "Bootstrap inference for inequality, mobility and poverty measurement," Journal of Econometrics, 108, 317-342.
Cameron C. and P. Trivedi (2005): Microeconometrics, Cambridge University Press.
Clarke J, Roy N. 2012. On statistical inference for inequality measures calculated from complex survey data. Empirical Economics 43: 499-524.
Cowell F. (2011) Measuring inequality, Oxford University Press.
Davidson R, Flachaire E. 2007. Asymptotic and bootstrap inference for inequality and poverty measures, Journal of Econometrics.
Davison A and D. Hinkley (1997): Bootstrap methods and their application, Cambridge University Press
Horowitz J.L. (2001). The Bootstrap. In: Heckman, J.J., Leamer, E.E. (Eds.), Handbook of Econometrics, vol. 5. Elsevier Science.
O’Donnell O., E. van Doorslaer, A. Wagstaff, M. Lindelow (2008): Analyzing Health Equity Using Household Survey Data: A Guide to Techniques and Their Implementation, The World Bank. Washington, D.C.
Schluter, C., & van Garderen, K. J. (2009). Edgeworth expansions and normalizing transforms for inequality measures. Journal of Econometrics, 150(1), 16-29.
Continue with Facebook
