[ugrads] Special Seminar Tuesday, February 21 - Miodrag Lovric

Myrna Walker mwalke13 at iit.edu
Thu Feb 16 14:13:44 CST 2017


Dear All,



Please join the Department of Applied Mathematics for a

*Special Seminar* - “This Talk Assumes only Univariate Calculus”



When: Tuesday, February 21, 2017, 12:45 pm - 1:50 pm

Where: RE 104



*Title: **Fueling Innovation: On the Deficiencies and Controversies of
Statistical Testing in Science*



*Speaker:* *Miodrag Lovric*



*Editor-in-Chief*

*International Encyclopedia of Statistical Science*





*Abstract:* Testing point null hypotheses is arguably the most widely used
and at the same time the most controversial, misapprehended and severely
criticized statistical procedure in many fields of scientific research. The
use and abuse of classical Null Hypothesis Significance Testing (NHST) have
been debated extensively for almost 80 years. We provide conclusive and
extensive arguments to ascertain our thesis that the accumulated problems,
controversies, and confusion in statistical testing have induced profound
statistical crisis in science. It is an imperative that we, statisticians,
took some measures by dismissing these anomalies within the current
paradigm and rebuilt healthy foundations of statistical science. Based on
the joint research with Professor C. R. Rao (Rao-Blackwell theorem,
Cramér–Rao lower bound), a new paradigm in the Kuhn’s sense of statistical
testing will be suggested. It is established on the implications of
so-called Zero probability theorem (paradox). The implications of this
theorem will be explored on the Fisherian significance testing,
Neyman-Pearson hypothesis testing and Bayesian testing based upon the Bayes
factor. Zero probability theorem implicitly leads to the final dissolution
of the famous Jeffreys-Lindley paradox, large sample paradox enunciated by
Berkson, and Bartlett’s paradox. We give enough arguments to confirm that
this will not only eradicate most of the objections against frequentist
testing and breathe a new life into them, but also considerably reconcile
communication in inference between frequentist and Bayesian approaches.



*Keywords and phrases: Point null hypothesis, Fisherian significance
testing, Neyman-Pearson hypothesis testing, Jeffreys-Lindley paradox,
Berkson’s paradox, Bartlett’s paradox, Cromwell’s rule, Bayes factor,
Zero-probability theorem, New paradigm in statistical testing.*







Pizza will be served.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://math.iit.edu/pipermail/ugrads/attachments/20170216/14ec69d2/attachment.html>


More information about the ugrads mailing list