Contents
Course Description
Course Programme
Request an Enrolment Form Now
Terms and conditions
About PcGive
About Ox
Timberlake Consultants Ltd, the publisher and distributor of the OxMetrics programs, invite you to attend a three-day course in Central London, covering Panel Data Analysis using OxMetrics.
The Course - Recently, empirical research in economics and the social sciences in general, has been enriched by the availability of a wealth of new sources of data: cross sections of individuals observed over time. The availability of panel data has stimulated a rapid growth in both methodological approaches and applications during the last 20 years. What is generally referred to as
panel data approach
to economic research provides several major advantages over conventional cross-sectional or time-series approaches. This facilitates the construction and testing of more realistic behavioural models that could not be identified using only a cross section or a single time-series data set.
Objectives of the Course - The objectives of this course are to provide participants with an introduction to panel data models, with the principle emphasis on empirical applications. This introduction progresses from standard 2 period models, fixed and random effects in static models, to dynamic homogeneous and heterogeneous models. Throughout the course an exposition of the underlying econometric issues will be given where appropriate.
Although this course makes use of menu driven procedures which may be used to estimate quite sophisticated panel models, instruction is also provided in the programming of relatively simple panel data estimators such as the Feasible Generalised Least Squares approach to Random Effects estimation. This will provide both an introduction to the flexible suite of matrix operators available in Ox, and also, by mirroing the statistical form of these estimators, will provide additional understanding that may not be obtained through menu based procedures.
Advantages - The course will
Applications - Throughout the course we use a wide range of economic applications. These will include:
- Models of Convergence: Application - China Provinces Data 1970-1995
- Currency and Banking Crisis: Application - Binary Response Models of Financial Crisis covering the period 1970-2000
- Models of Manufacturing Output: US Data for 1979-1988
Who Should Attend - The course is designed to attract both policy and academic analysts working with panel data. Whilst the instruction makes no assumptions as to prior exposure to panel data methods, participants will require a basic familiarity with standard econometric techniques, such as multiple regression, instrumental variables, and various estimation methods such as maximum likelihood and the generalised method of moments. Familiarity with basic programming concepts and empirical analysis will also be advantageous but not essential.
The Principal Lecturer - The principal lecturer is:
Dr. Melvyn Weeks, University Lecturer in Economics at the University of Cambridge, U.K.
Cost - The cost of the course are:
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.
Day 1 - Using Ox and PcGive for Panel Data
Session 1: Introduction to Ox
- Why use Ox
- Introduction to the OxMetrics programs
- An overview of programming in Ox including the Object Oriented end approach
- Estimation and Inference Using Ox and PcGive
Session 2: Practical Session
- Data input and manipulation
- A simple OLS estimator in matrix form: A first Ox program
- Creating and Manipulating panel datasets: key Ox functions for panel data manipulation including vec(), shape(), diff0(), and the Kroenecker Product
- Calling libraries in Ox
- Using PcGive
Session 3: An overview of Panel Data Estimators
- An overview of Panel Data Estimators
- A simple two period model
- Linear Unobserved Effects Panel Data Models
- Fixed, Random Effects, and the First Difference Estimator
Day 2 - Using Ox and PcGive for Panel Data (cont.)
Session 1: Practical Session
- Feasible Generalised Least Squares (FGLS) estimation of Random Effects models
- Matrix programming in Ox to obtain the FGLS estimator for heteroscedasticity and serial correlation in idiosyncratic error
- Fixed and First Difference Estimators for Policy Analysis
Session 2: Dynamic Panel Data
- Dynamic unobserved effects models
- Attenuation bias in the standard dynamic panel data model
- An overview of estimator approaches: Anderson and Hsiao (1982), Blundell and Bond (1998), Arelano and Bover (1995)
- Homogeneous versus Heterogeneous slopes
- The Initial Condition Problem
Session 3: Practical Session
- An introduction to Panel Data Estimators using PcGive and Ox
Session 4: General Methods of Moments (GMM)
- An overview of GMM models
Day 3 - Practical Estimators of Multi-Regional Analysis of Growth
Session 1: Practical Estimators of a Multi-Regional Analysis of Growth
- GMM Estimators of a Multi-Regional Analysis of Growth
- The Problem of Weak Instruments
- GMM estimation of a growth model for the transition economies
Session 2: Miscellaneous Topics
- The Mean Group and Pooled Mean Group Estimators
- Panel Data Estimators and Latent Variable Models
Session 3: Panel Data Estimates of Binary Response Models
- Pooled Probit and Logit Models
- Unobserved Effects Probit Models
- A Random Effects Probit Model with an application to Models of Currency Crisis
