Forecasting and Analysing Time Series using Kalman filter methods within OxMetrics™

3-5 April 2007

The George Washington University, 2121 Eye Street, N.W. Washington, D.C. 20052 A.

Contents

Course Description
Course Programme
Request an Enrolment Form Now
Terms and Conditions
About STAMP

Timberlake Consultants Ltd, the distributor and publisher of the OxMetrics™ software, would like to invite you to attend a three-day course, in Washington DC. This course will discuss the ideas of analyzing, modeling and forecasting time series using Structural Time Series Models and more general unobserved components time series models. The course will give an in-depth treatment of this methodology of time series analysis and how it can be implemented using the OxMetrics™ module STAMP™ and the SsfPack™ packages.

Participants may want to combine this course with one or more of the following courses/events:

The course - Several major advances in time series, forecasting and software engineering have occurred in the past years. These advances have provided a major breakthrough in the modelling of time series using easy-to-use object-oriented Windows-based software. This course aims to provide participants with a thorough understanding of structural time series models, unobserved components, state space, the Kalman filter, signal extraction and forecasting. It further demonstrates how this innovative modeling and forecasting methodology can be implemented in real-life business, industry and government situations. The course also pays attention on how to interpret and report results using the STAMP™ and SsfPack™ software packages. However, you do not need to be a STAMP™ user. Developers of other software packages, e.g. EViews™, ForecastPro, have followed the work done by the STAMP™ and SsfPack™ developers when implementing this type of models. Participants are invited to send their own data in Excel format prior to the start of the course.

Who should attend? - Structural time series models find application in many subjects, including, economics, finance, sociology, management science, biology, geography, meteorology, transportation, tourism and engineering. The course is, therefore, suitable for anyone who works with time series data in business, industry and government or for those who teaches time series at universities. Only a basic knowledge in regression and time series analysis is assumed. Familiarity with the software is not required. If you want to get some familiarity with the software prior to the course, please request a demo copy.

The Principal Lecturer - The principal lecturer is:
Prof. Siem Jan Koopman is a Professor in Econometrics at the Vrije Universiteit Amsterdam. He gained his Ph.D. (from LSE) in 1992 and worked earlier at the LSE and Tilburg University . He has published articles in Biometrika, JASA, J Business and Economics Statistics, and J Royal Statistical Society - Series B. He is co-editor of Econometrics J and J of Forecasting and is editorial board member of J of Applied Econometrics. He is the main contributor to the OxMetrics™ module STAMP™ and developed the Ox package SsfPack.

http://stamp-software.com
http://www.ssfpack.com

Cost - The cost of the course is:

Organization Type
1st Participant
2+ Participants
Commercial
$2,495.00
$2,245.00
Government
$1,900.00
$1750.00
Academic
$1,500.00
$1,500.00

Discounts

  • 20% discount for attending 2 courses
  • 30% discount for attending 3 courses
  • 10% for holders of the OxMetrics Enterprise Edition
  • a further 5% discount is provided to subscribers of the Foresight magazine

The cost includes course materials, course dinner, lunch, refreshments and the use of computers. The number of delegates is restricted. Please register early to guarantee your place. Further instructions will be sent with the joining instructions. If you need assistance in locating hotel accommodation in the area, request the help of our Training Department.

Top

Agenda
(subject to minor changes)

Day 1

9.30 Coffee and registration

10.00 Welcome

Introduction to Unobserved Components Time Series Models:

univariate models * statistical properties * connection with ARIMA * trends * unobserved components

Hands-on session with STAMP

Lunch

Signal Extraction and Seasonal Adjustment:

weights * filters * forecasting * seasonal adjustment * business cycles

Hands-on session with STAMP

17.00 Close

Day 2:

9.30 Start

Introduction to Kalman filter and state space models:

local level model * filtering * prediction * smoothing * simulation * forecasting * missing values

Discussion, review and some exercises

Hands-on session with STAMP

Lunch

Analysing and Modeling Time Series:

decompositions * regression effects * detection of outliers and breaks

Analysing and Modeling Multiple Time Series:

multivariate models * common factors * common trends and cointegration * common cycles * VARs * VECMs

17.00 Close

Day 3

9.30 Start

State space methods and the Ox module SsfPack:

Kalman filter and related algorithms * likelihood evaluation * estimation * residuals * diagnostic checking

Hands-on session with Ox/SsfPack

Lunch

Applications in economics and finance using model-based approaches:

forecasting * measuring trends and business cycles * stochastic volatility * realised variance

Recent Advances in State Space and Unobserved Components

Review and discussion

17.00 Close

Terms and Conditions

Registration closes 5 calendar days prior to the start of the course.

Cancellations:

  • full fee returned for cancellations made over 28 calendar days prior to start of the course
  • half-fee returned for cancellations made 14 calendar days prior to he start of the course
  • no fee returned for cancellations made less than 14 calendar days prior to the start of the course.

    Payment of course fees required prior to the course start date

For Timberlake Consultants Terms and Conditions click here