My name is Bera, Jarque-Bera


Greg Mankiw recently wondered whether he is a typo. He shouldn't be complaining; the man in the picture, on the other hand, has every right to.

Bluematter. is proud to award Carlos M. Jarque the Distinguished Medal for Most Misspelled Economist of All Times. On Google Scholar, a search for Jacque-Bera (see note to editors at the bottom of this post) returns no less that 177 results, while a search for 'Jacque Bera normality' generates 581. Many of these papers are published in prestigious journals, and my brief inquiry revealed a handful of papers (e.g.) which include tables that give the value of the Jarque-Bera statistic while referring to the Jacque-Bera statistic in the main text. Analysis suggests that the Jarque Bera statistic is misspelled between 5% and 15% of the times in published papers. Further anecdotal evidence points to less than 20% of the people getting the true name of the JB statistic right in conversation and unpublished work. To (roughly) quote Indifference Merv: 'It makes it difficult to figure out whether there are in fact two distinct statistics, Jarque-Bera and Jacque-Bera. Jarque has been done a great injustice by the profession'.

Note to Editors: Carlos M. Jarque is an economist with a long and distinguished career in economics, politics and management. Amongst econometricians, he is best known for his contribution (with Anil K. Bera) to testing for normality of observations and regression residuals: the Jarque-Bera statistic (here is wikipedia, and here is the paper)

Postscript: Anil K. Bera does not fare much better. A search for Jarque-Berra returns 173 papers.



by datacharmer | Monday, October 01, 2007
  | | My name is Bera, Jarque-Bera @bluematterblogtwitter

0 comments: