Timberlake Software

TBA
Department of Economics, Manor Road Building, Manor Road, Oxford, OX1 3UQ, UK

Contents

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

Timberlake Consultants Ltd, the publisher and distributor of OxMetrics, invite you to attend a four-day course in Oxford. The course concerns the theory and practice of econometric modelling and forecasting in a non-stationary and evolving world, when the model differs from the economic mechanism. The course has been designed for economic researchers in Central Banks and Ministries of Finance.

This is the second opportunity to attend such course. We run this course one year ago and its success has driven us to run it again. The course has been designed by Prof. David F. Hendry, who is also one of the main lecturers, and we cannot commit to be able to repeat it.

Prerequisite - A reasonable background in basic econometric methods and knowledge of econometric concepts relevant for time-series data are essential to benefit from the course. Some experience in undertaking empirical research would be an advantage. Laptops will be needed for the computer labs.

The Course

Econometric Modelling and Forecasting

The course concerns the theory and practice of econometric modelling and forecasting in a non-stationary and evolving world, when the model and mechanism differ. The main model class is a vector autoregression in integrated-cointegrated variables leading to an equilibrium-correction system, but intermittently subject to structural breaks. The framework, its basic concepts and main implications will be sketched, based on the theory of reduction, which explains how the data generation process (DGP) leads to models thereof. Models with no losses on reduction are congruent; those that explain rival models are encompassing. The main reductions correspond to key econometrics concepts (causality, exogeneity, invariance, etc.), and are the null hypotheses of model-evaluation tests, shown by a taxonomy of evaluation information. Congruent and encompassing sub-models can, therefore, be justified, so we consider how they should be selected.

Model selection

Theory is difficult: all statistics for selecting and evaluating models have interdependent distributions, altered by every modelling decision. Recent research on model selection for general-tospecific(Gets) modelling will be described, emphasizing automatic procedures. Gets mimics reduction by simplifying a congruent general unrestricted model to a dominant minimal representation. Computer automation of selection algorithms has revealed high success rates, and allows operational studies of alternative strategies. Developments to Autometrics are explained. We consider its performance across different states of nature, and examine the extent to which model selection is non-distortionary at relevant sample sizes. Collinear data problems; indicator saturation methods and their generalization to more candidate variables than observations; and tests for, and modelling of, non-linearity and invariance will all be described.

When the processes being modeled are not time invariant, many of the famous theorems of economic forecasting no longer hold; rather their converses often do: e.g., non-causal devices may outperform causal. Six aspects of unpredictability in forecasting compound the four additional mistakes most likely in empirical models. A generalized taxonomy of forecast errors is developed, allowing for structural change in the forecast period, the model to be mis-specified for the DGP over the sample period, and selected from sample evidence, the parameters of the model to be estimated (possibly inconsistently) from the data, which might be measured with error, the forecasts to commence from incorrect initial conditions, and innovation errors to cumulate over the forecast horizon. The taxonomy reveals the central role of unanticipated location shifts, and helps explain the outcomes of forecasting competitions. Other potential sources of forecast failure seem less relevant. Corrections to reduce forecast-error biases (intercept and forecast-error corrections), model transformations (differencing), pooling, and rapid updating are all shown to help robustify forecasts in the face of location shifts.

The afternoon sessions are devoted to computer exercises, commencing with introductions to OxMetrics and PcGive (including Autometrics) as basic modelling tools, including cointegration and system procedures. Applications of Gets, Monte Carlo simulation (using PcNaive), and empirical forecasting using PcGive all illustrate the theory sessions. Although OxMetrics is the software being used during the practical sessions, this course does not aim to teach OxMetrics. The course aims to teach the theory and practice of econometric modelling and forecasting in a non-stationary and evolving world and will benefit users of any computer software system.

Readings :

[1] Hendry, David F., and Grayham E. Mizon, 2000, “Reformulating
Empirical Macro-econometric Modelling,” Oxford Review of Economic Policy, Vol. 16 (Winter), No. 4, pp. 138–59. http://eprints.soton.ac.uk/33098/

[2] Hendry, David F., 1995, “The Theory of Reduction,” in Dynamic Econometrics ( Oxford : Oxford University Press), Chapter 9 http://www.ingentaconnect.com:80/content/oso/611411/1995;jsessionid=2ne103rptm77.alexandra

[3] –––––– and Hans-Martin Krolzig, 2005, “The Properties of Automatic Gets Modelling,” Economic Journal, 115, C32–C61.
http://www.blackwell-synergy.com/doi/abs/10.1111/j.0013- 0133.2005.00979.x

[4] Hoover , Kevin D., and Steven J. Perez, 1999, “Data Mining Reconsidered: Encompassing and the General-to-Specific Approach to Specification Search,” Econometrics Journal, Vol. 2, pp. 167–91.
http://www.blackwell-synergy.com/doi/abs/10.1111/1368-423X.00025

[5] Clements, Michael P., and David F. Hendry, 2003, “An Overview of Economic Forecasting,” in Companion to Economic Forecasting, ed. by Clements and Hendry ( Oxford : Blackwells), pp.1–18.
http://www.blackwellreference.com/subscriber/uid=959/tocnode?id=g97814 05126236_chunk_g97814051262363

[6] ––––––, 2005, “Overview to Information and Model Transformations in Economic Forecasting,” Oxford Bulletin of Economics and Statistics, 67, 713–753
http://www.blackwell-synergy.com:80/doi/abs/10.1111/j.1468-0084.2005.00139.x

[7] Hendry, David F., 1997, “The Econometrics of Macroeconomic Forecasting,” Economic Journal, Vol. 107, 1330–57.
http://www.jstor.org:80/view/00130133/di983548/98p0438z/0

Background Readings Clements, Michael P., and David F. Hendry, 1998, Forecasting Stationary Economic Time Series ( Cambridge University Press).––––––, 1999, Forecasting Non-Stationary Economic Time Series (MIT Press).

Doornik, Jurgen A., and David F. Hendry, 2007, Interactive Monte Carlo Experimentation in Econometrics using PcNaive (Timberlake Consultants Press).

Doornik, Jurgen A., and David F. Hendry, 2007, Empirical Econometric Modelling using PcGive I (Timberlake ConsultantsPress).

Hendry, David F., and Neil R. Ericsson, eds., 2001, Understanding Economic Forecasts (MIT Press).

Hendry, David F., and Bent Nielsen, 2007, Econometric Modeling: A Likelihood Approach ( Princeton University Press).

Who should attend - The course, given in English, is aimed at forecasters and researchers working for

Central Banks
Ministries of Finance

Advantages - The course will

Review all major econometric modelling methods
Provide a practical and systematic approach to the modelling and forecasting in a non-stationary and evolving world
Provide hands-on experience in building econometric models - each delegate is provided with a computer throughout the course

The Principal Lecturers - The principal lecturers are:

Prof. David F. Hendry (http://www.nuffield.ox.ac.uk/users/hendry/hendry.html)
Dr. Jennifer F. Castle (http://www.economics.ox.ac.uk/students/jennifer.castle/home.htm)

Timetable

Wednesday 26 March

5:30-6:30	Registration Keble College, Porters Lodge
7.00	Welcome reception TBA

Thursday 27 March

9:00-9:30	Coffee Department of Economics
9:30-11:00	Session 1 General introduction to modelling and forecasting
11:00-11:30	Coffee Break
11:30-1:00	Session 2 Introduction to OxMetrics ‘front end’
1:00-2.00	Lunch
2.00-3.30	Session 3 Introduction to model selection
3.30-4.00	Tae Break
4.00-5:30	Session 4 Introduction to PcGive

Friday 28 March

9:00-9:30	Coffee Department of Economics
9:30-11.00	Session 5 Theory of model selection
11:00-11:30	Coffee break
11:30-1:00	Session 6 Introduction to Autometrics
1:00-2:00	Lunch
2:00-3:30	Session 7 Introduction to economic forecasting
3:30-4:00	Tea break
4:00-5:30	Session 8 Introduction to PcNaive
7:00	Course Dinner TBA

Saturday 29 March

9:00-9:30	Coffee Department of Economics
9:30-11:00	Session 9 Potential problems with forecasting and some solutions
11:00-11:30	Coffee break
11:30-1:00	Session 10 Forecastin in PcGive
1:00-2:00	Lunch
2:00-3:30	Session 11 Question and answer session

All lunches and coffee/tea breaks will be held in the Department of Economics common room on the 1st floor of the Manor Road building.
All teaching will take place in the computer room on the 2nd floor of the Department of Economics, Manor Road Building.

Cost - It is generally difficult to find accommodation in Oxford. We have, therefore, reserved some rooms at Keeble College. The cost of the course is: