A Study of the Estimation of the Gini Coefficient of Income Using Lorenz Curve
Kwasi A. Darkwah *
Department of Statistics, School of Physical and Mathematical Sciences, College of Basic and Applied Sciences, University of Ghana, Ghana.
Ezekiel N. N. Nortey
Department of Statistics, School of Physical and Mathematical Sciences, College of Basic and Applied Sciences, University of Ghana, Ghana.
Felix O. Mettle
Department of Statistics, School of Physical and Mathematical Sciences, College of Basic and Applied Sciences, University of Ghana, Ghana.
Isaac Baidoo
Department of Statistics, School of Physical and Mathematical Sciences, College of Basic and Applied Sciences, University of Ghana, Ghana.
*Author to whom correspondence should be addressed.
Abstract
Aims: This paper compares the Boole and Weddle numerical integration methods to estimate the Lorenz curve and Gini Coefficient of income in Ghana.
Study Design: Research Paper.
Place and Duration of Study: Ghana, Secondary data for 2013 Ghana Living Standard Survey.
Methodology: The Lorenz curve and Gini coefficients of income were estimated using Rasche, Gaffney and Obst function and polynomial function according to numerical integration methods such as Boole and Weddle methods. The Bias and relative error was used to compare the numerical integration methods used.
Results: The results showed that the estimated Lorenz curve and Gini coefficients using Rasche, Gaffney and Obst function and polynomial function according to the Boole and Weddle method of integration resulted in positive and negative biases respectively with the Boole method producing the highest absolute relative error of 1.8082%.
Conclusion: This study showed that both the Boole and Weddle method of numerical integration are not uniformly optimal in estimating the Gini coefficient of income but the Weddle’s method is better as compared to Boole method of numerical integration in estimating the Gini coefficient of income.
Keywords: Boole, Weddle, lognormal, income, Lorenz curve, Gini coefficient