Thoung Thi Mai Nguyen

Thoung's paper on distribution-free goodness of fit test in in contingency tables was just publoshed in Metrica (2016) . She showed that one can go far beyond usual chi-square test.

The class of tests, which is possible to use now is very broad. In comparison to my 2014 paper, Thuong demonstrated that possible high-dimensionality of the parameter -- for example in 6x7 table there are 30 parameters -- the method works quickly and reliably, and the convergence is still quite good.

She is close to finish her second paper on disribution-free goodness of fit approach for testing if the tail of the distribution is regularly varying. This approach is easier to implement than the one in Can,Einmal,Khmaladze,Laeven(AS,2014), but it is not yet described in $R^d$.

This topic: Users/Estate > CurrentPhDStudents
Topic revision: 14 Dec 2016, estate
Contact Us | Section Map | Disclaimer | RSS feed RSS FeedBack to top ^

Valid XHTML and CSS | Built on Foswiki

Page Updated: 14 Dec 2016 by estate. © Victoria University of Wellington, New Zealand, unless otherwise stated