Deirdre McCloskey in DC
Posted by Evan Herrnstadt on November 25, 2008
From the inbox:
Upcoming Luncheon of the Society of Government Economists
Thursday, December 4, 2008 (12:00 – 1:30 PM)
“Statistical Significance is Essentially Meaningless (at the 5% Level).”
Deirdre McCloskey and Stephen Ziliak have been arguing for some years that accepted practices in econometrics are mistaken. In their current book, The Cult of Statistical Significance: How the Standard Error Costs Jobs, Justice, and Livesthey present the full case, drawing on examples from medicine, psychology, and economics. They note that the very simple point is not original.It has been made repeatedly by hundreds of the best statistical theorists and practitioners (Savage, Friedman, Zellner, Kruskal, Heckman) over the past century. And yet users of statistics continue to kill patients and misadvise governments. Why?
Chinatown Garden Restaurant, 618 H St., NW, Washington, DC
(For reservations, contact Steven.Payson@bea.gov)
Reservations are $16 for members, $20 for non-members.
You might recall that I posted on this thread of McCloskey’s research a few months ago. Just in case you want to relive my econometric frustrations prior to the lunch.