The Small-Sample Bias of the Gini Coefficient: Results and Implications for Empirical Research

By George Deltas. Review of Economics and Statistics, 2003

271 citations recorded by the source corpus.

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

Reduced, Reused and Recycled: The Life of a Dataset in Machine Learning Research (2021, methodology): “Because Gini can be biased in small samples [36], we use the the sample-corrected Gini,Gs = n n−1G, 199 and excluded tasks (or task-years when disaggregating by time) with fewer than 10 papers.”
Trade-offs between social and environmental Sustainable Development Goals (2018, methodology): “…AL, especially with small sample sizes ( =n 4), and, as such, underestimateG, we reduce the bias by calculating an adjusted Gini index (Gadj; Deltas, 2003): =G n n G 1adj As an alternative measure of inequality, we also calculate the coefficient of variation, i.e. the ratio of the standard…”
Reliable Inference for the Gini Index (2009, unknown): “A more recent contribution is Deltas (2003), where the source of the bias of Ĝ is given a geometric interpretation, and its link to the two definitions of the finite-sample index.3 In view of all this, it is a little strange that the definition (42) given by Sen (1976) of his poverty index does…”
Cross-cultural, developmental psychology: integrating approaches and key insights (2020, background): “For instance, while the Gini coefficient that can indeed be used to predict variation in behavior, such as people's preferences for redistribution (Kerr, 2014), it is also sensitive to a number of biases such as the small-sample bias — see (Deltas, 2003).”