[Statlist] talks on statistics

Christina Kuenzli kuenz|| @end|ng |rom @t@t@m@th@ethz@ch
Mon Nov 26 13:58:03 CET 2007


                  ETH and University of Zurich 

                           Proff. 
         A.D. Barbour - P. Buehlmann - F. Hampel - L. Held
                  H.R. Kuensch - S. van de Geer

***********************************************************    
           We are glad to announce the following talks
***********************************************************

      Thursday, November 29, 2007, 16.15h - 17.00 LEO C 6

      Adaptivity of the monotone least squares estimator
      Eric Cator, Delft University of Technology, NL

In this talk we will consider the estimation of a monotone 
regression (or density) function in a fixed point by the least squares 
(Grenander) estimator. We will show that this estimator is fully 
adaptive, in the sense that the attained rate is given by a functional 
relation using the underlying function $f_0$, and not by some 
smoothness parameter, and that this rate is optimal when considering 
the class of all monotone functions, in the sense that there exists a 
sequence of alternative monotone functions $f_1$, such that no other 
estimator can attain a better rate for both $f_0$ and $f_1$. When 
defining the rate, we do not look at the expectation of some convex 
loss function, but rather we bound the probability that the difference 
between the estimator and the true value is larger than the given rate 
(probabilistic error).

***********************************************************

      Friday, November 30, 2007, 15.15h - 17.00 LEO C 6

      Building Bridges: Continuous-time imputation from financial time
      series to Longitudinal Data Analysis
      Don McLeish University of Waterloo, Ontario

Diffusion processes are very popular for the pricing and hedging of
derivative securities and more generally for the analysis of
continuous time data.  We are often concerned with imputing missing or
latent  values in a time series or calculating  the expected value of
the functional of a diffusion process.  Examples of such functionals
include quantiles or first passage times (as in risk management),
conditional expectations, or the distribution of various test
statistics.
When properties are available for the simplest of continuous-time
processes such as Brownian motion, we will show how simulation
techniques allow us to port these to more complex diffusion processes
using unbiased importance sampling.
We discuss various applications to pricing exotic options and to the
analysis of longitudinal data.      
________________________________________________________
Christina Kuenzli            <kuenzli using stat.math.ethz.ch>
Seminar fuer Statistik      
Leonhardstr. 27,  LEO D11      phone: +41 (0)44 632 3438         
ETH-Zentrum,                   fax  : +41 (0)44 632 1228 
CH-8092 Zurich, Switzerland        http://stat.ethz.ch/~




More information about the Statlist mailing list