A check of Maddison’s gdp data. Benford’s Law with some range problems
The paper studies if the Maddison set of data for GDP per capita follows a statistical regularity, known as Benford’s Law. It is a simple logarithmic relation on the frequency of the first digit in a data set. These data ought to follow the law as they are Maddison’s calibration of data compiled by many independent agencies and researches. The data set consists of 12,411 observations, permitting a rather strong test. On visual inspection the relation appears to fit rather well, but the law is rejected by a formal test. The explanation of the rejection is found to be that the range of the data is too small.
When requesting a correction, please mention this item's handle: RePEc:aah:aarhec:2010-18. See general information about how to correct material in RePEc.
For technical questions regarding this item, or to correct its authors, title, abstract, bibliographic or download information, contact: ()
If references are entirely missing, you can add them using this form.