Workshop on Asymptotic Methods in Data Science

Supported by Victoria University of Wellington and Keio University

Organisers: Yuichi Hirose and Ritei Shibata

2011-11-23

The objectives of this workshop are not only to provide an opportunity for the exchange of research ideas and results, but also to accelerate cooperative research on asymptotic methods and their application in data science. The overall aim is the better modelling of the various phenomena that underpin data drawn from the natural and social sciences, among others. To encourage research interaction and developments, the workshop programme is designed to be flexible so that it can accommodate discussion and suggestions received from participants.

Date: 2011-12-06 (Tue) ~ 2011-12-08 (Thu)

Place: Discussion Room 7 (14-217), Yagami Campus, Keio University

Programme

Dec. 6

Ÿ Lecture“Introduction to goodness of fit problems and statistical practice 1” by Estate Khmaladze (VUW)

10:30-12:00 Why problems with distribution free testing occur and why innovations can solve them; simple examples of innovations even for discrete time

12:00-14:00 Lunch break

Ÿ Session 1: Asymptotics in parametric inference

14:00-15:00 Using a mixture of estimation methods for fitting the GEV distribution by Peter Thomson (SRA)

15:00-16:00 Asymptotics of spatial AR parameter estimation by Yuki Takei and Ritei Shibata (Keio)

16:00-16:30 Tea/Coffee break

Ÿ Session2: Asymptotic distributions

16:30-17:30 Asymptotic distribution of random partition by Masaaki Sibuya (Keio)

18:00-20:00 Welcome Party at Forum, Yagami Campus

Dec.7

Ÿ Lecture “Introduction to goodness of fit problems and statistical practice 2” by Estate Khmaladze (VUW)

10:30-12:00 Goodness of fit test; multitesting problem

12:00-14:00 Lunch break

Ÿ Session 3: Asymptotics of semi-parametric models

14:00-15:00 Information criteria for semi-parametric models by Yuichi Hirose (VUW)

15:00-16:00 Asymptotics of semi-parametric models by Ryozo Miura (Hitotsubashi)

16:00-16:30 Tea/Coffee break

16:30-17:30 Asymptotics for L-statistics with dependent data and applications to risk measurement by Hideatsu Tsukahara(Seijo)

17:30-18:30 Semiparametric penalized spline regression by Takuma Yoshida and Kanta Naito (Shimane)

Dec. 8

Ÿ Lecture “Introduction to goodness of fit problems and statistical practice 3” by Estate Khmaladze (VUW)

10:30-12:00 Innovations (KhT) in regression

12:00-14:00 Lunch break

Ÿ Session 4: Asymptotics of goodness of fit test statistics

14:00-15:00 Approximation of Cramer-von Mises test statistic by weighted sum of chi-square variables by Mayumi Naka and Ritei Shibata (Keio)

15:00-16:00 Possibilities of distribution free goodness of fit tests for discrete distributions by Estate Khmaladze (VUW)

16:00-16:30 Tea/Coffee break

Ÿ Overall Discussion

19:00-21:00 Dinner Party at Faculty Lounge, Hiyoshi Campus

Contact: shibata@math.keio.ac.jp