Summer School on Modern Statisitical Analysis and Computational Methods

focusing on Time Series Analysis

Professor Dimitris Kugiumtzis; Mr Liam Kilmartin (Aristotle University of Thessaloniki; Greece; NUI Galway)

Cairnes Building, NUI Galway 09:00 :: 17th June 2013

Social Sciences Computing Hub at the Whitaker Institute and HRB’s Clinical Research Facility jointly host the first Summer School on Modern Statistical Analysis and Computational Methods at NUI Galway from June 17-19th, 2013

This year's school will focus on “Time Series Analysis” with applications in a vast array of disciplines ranging from Physiology to Finance. The presentation will be lecture cum hands-on computer simulation using empirical data from relevant application domains.

The School will run for three days. Professor Dimitris Kugiumtzis from Aristotle University of Thessaloniki, Greece will deliver the first two days, and Mr. Liam Kilmartin from the College of Engineering and Informatics at NUI Galway will conduct the third day’s workshop.


Workshop I : June 17th and 18th, 2013
Time Series Analysis: Applications in Physiology, Climate Change and Finance

Presenter: Professor Dimitris Kugiumtzis, Aristotle University of Thessaloniki, Greece.

Duration: This two-day course will run from 9:00am to 5:00pm on Monday 17th & Tuesday 18th June.

Course Content:

Basic characteristics of time series: Stationarity; autocorrelation; removal of trends and seasonality; independence test of time series.

Linear stochastic models for time series: Autoregressive (AR), moving average (MA), autoregressive moving average (ARMA) models for stationary time series; autoregressive integrated moving average (ARIMA) models for non-stationary time series; models for seasonality in the time series; prediction of time series.

Nonlinear analysis of time series: Extensions of linear stochastic models; nonlinear characteristics of time series; nonlinear dynamics and chaos; nonlinear prediction of time series.

Case studies on time series analysis: detecting structural changes in time series records; examples from physiology and finance. Analysis using a stand-alone GUI software.


About Professor Kugiumtzis

Dimitris Kugiumtzis (BSc in Mathematics, MSc and PhD in Informatics) is Associate Professor at the Department of Mathematical, Physical and Computational Sciences, Aristotle University of Thessaloniki, Greece. His main research area is time series analysis in conjunction with dynamical systems and chaos, as well as computational statistics and data mining. Applications extend from physiology to climate and finance. He has published over 40 journal papers and many national and international proceedings papers. He has participated in several national and European research projects, acted as EU evaluator and he is regular reviewer for several journals. He has supervised 4 completed PhD and many MSc.


Workshop II : June 19th, 2013
Time Series Analysis: Signal Processing Methods

Presenter: Liam Kilmartin, NUI Galway

Duration: 9:00 -17:00

Course content:

Basic Time Domain Concepts in Signal Processing: Singal sampling, Fourier decomposition of time domain signals, difference equations, Linear digital filter, filter design, IIR and FIR filters, Periodicity estimation

Frequency Domain Analysis: Fourier analysis, DFT, Impact of DFT window shape and size, Power Spectral Density, Stationarity and frequency domain analysis, revisiting AR(MA) modeling from a frequency domain perspective, introduction to advanced topics such as wavelet analysis, linear coherencesynchrony

Signal Processing Applications: Application of basic signal processing techniques in the fields of physiology, finance and mobile phone based sensing for social science based research; Examples will use Matlab and Android mobile app technology.

About Mr. Kilmartin

Liam Kilmartin received his BE (Electronic Engineering) degree in 1990 and his MEngSc degree in 1994, both from NUI, Galway. He joined the academic staff of the department in 1992 as a temporary teaching appointment and he was appointed to a permanent position in 1994 specialising in the fields of fixed and mobile communication systems. His current research interests focus on the application of signal processing and machine learning techniques in a variety of domains. These include speech, audio and image processing, mobile and fixed network modelling, signal processing and classification algorithms for EEG analysis with specific interest in Brain Computer Interface applications, image processing for biomedical imaging modalities and the application of signal processing to bio-signal analysis in general. He has supervised students to both masters and PhD awards in these areas, funded through IRCSET scholarships and competitive external funding which he has individually attracted, and he has a significant number of publications in international journals and peer reviewed conference proceedings.



Number of places: 30

For further queries, contact Gwen Ryan at gwen.ryan@nuigalway.ie

Registration required

Links

Registration

 
ISSP Sponsors: