THE ROYAL STATISTICAL SOCIETY
=================================
TIMSAC (Time Series and Applications) Study Group Meeting
-------------------------------------------------------------------
Thursday, 18 June 1998, at 4 p.m.
==========================================
The Royal Statistical Society, 12 Errol Street, London EC1
(Nearest Underground stations: Barbican, Moorgate)
OTTAR N. BJORNSTAD
(University of Oslo, Norway)
" Stochastic population dynamics and time series analysis of abundance:
two case studies "
CEES DIKS
(University of Kent)
" Estimating invariants of noisy attractors "
A L L A R E W E L C O M E !
SUMMARIES
OTTAR N. BJORNSTAD:
The abundance of animals is governed by nonlinear interactions
between individuals and external forcing. The environmental forcing
and the birth and death of individuals is typically stochastic.
Abundance data thus represents observations from a nonlinear and
non-Gaussian process. I talk about time series analysis of
population fluctuations in two case studies using a nonparametric
(functional coefficient) model and a parametric random-coefficient
model. In addition to the statistical modelling, the goal is to
illustrate the process of ecological inference from time series
data.
Tea break at 4:45 p.m.
CEES DIKS:
We propose a method based on Gaussian kernel correlation integrals
for estimating the correlation dimension, correlation entropy and
noise level of chaotic time series with Gaussian observational noise.
For simulated data good results are obtained even for considerable
noise levels. We present several applications to experimental time
series.
-----------------------------------
Information on TIMSAC can be found on:
http://www.zoo.cam.ac.uk/timsac/timsac.htm
or contact:
B"arbel Finkenst"adt, Department of Zoology, University of Cambridge,
Downing Street, Cambridge CB2 3EJ, [log in to unmask]
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
|