Venue:

Subotnick Financial Services Centre, Zicklin School of Business, Baruch College/CUNY
Information and Technology Building, 151 E. 25th Street. New York, NY 10010, U.S.A.

The Course - Several major advances in time series, forecasting and software engineering have occurred in the past few years. These advances provide a major breakthrough in the modeling of time series using easy-to-use object-oriented Windows based software. The course overviews econometric modeling and uses real-life business and industrial data to show how to interpret and report the results. Participants are welcome to bring their own data.

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

* Economic Research/ Model Building
* Financial Modelling/ Arbitrage Trading
* Quantitative Investment Management
* Sales and Inventory Forecasting
* Traffic Modellers
* Energy Load Forecasting
* University Instruction
* and more

Advantages - The course will

* Review all major econometric modelling methods
* Provide a practical and systematic approach to the modelling of business and financial time series data
* Provide hands-on experience in building econometric models - each delegate is provided with a computer throughout the course
* Provide an opportunity for you to meet with a panel of experienced modellers to discuss industry-specific applications related to forecasting and time series analysis

Agenda
(subject to minor changes)

Day 1 - Using EViews for Time Series Forecasting

Session 1: Introduction to EViews

* Data Entry and Management
* Seasonal Adjustment of Time Series
* Modelling a Time Series – the Box-Jenkins Approach
* Generating and evaluating a forecast

Session 2: Practical Session

* Fitting a Box-Jenkins model
* Generating a forecast

Session 3: Stationarity, unit roots and forecasting

* Testing for unit roots
* Implications of unit roots for forecasting
* Unit roots and the spurious regression problem

Session 4: Practical Session

* Practical testing for unit roots
* Forecasting with non-stationary data

Day 2 - Regression Analysis Using EViews

Session 1: The Classical Linear Regression Model

* Methods of Estimation – Least Squares, Maximum Likelihood, Method of Moments
* Approaches to Testing – Wald, Likelihood Ratio and Lagrange Multiplier
* Diagnostic Testing Using EViews

Session 2: Practical Session

* Estimating a Model Using the General to Specific Approach
* Diagnostic Testing
* Testing restrictions

Session 3: Cointegration and error-correction

* Dealing with Cointegrated Variables
* The Engle-Granger Two Stage Procedure
* Johansen’s approach to testing for cointegration
* Cointegration and error-correction models

Session 4: Practical Session

* Testing for Cointegration
* Forecasting with an error -correction model

·Day 3 - Specialised Topics

Session 1: Panel Data

* Fixed and random effects models
* Unit roots in panel data
* Cointegration and panel data

Session 2: Practical Session

* Setting up a panel data model
* Estimation and interpretation of results with panel data

Session 3: Vector Autoregression Models and Exogeneity

* The VAR Methodology
* Granger causality
* Exogeneity

Session 4: Practical Session

* Testing for exogeneity
* Estimation and interpretation of a VAR
* Using a VAR to generate a forecast

Day 4 - Using EViews for Modelling

Session 1: Building a Model

* The Process of Estimating and Setting up a Model Within EViews
* Dynamic and Static Solutions of a Model
* Forecasting With a Model
* A Small Macroeconomic Model of the United States Economy

Session 2: Practical Session

* Setting up a Model and Generating a Forecast

Session 3: Economic Theory and Econometric Models

* The Importance of Long Run Theoretical Restrictions
* The Effects of a Non Unit Income Elasticity of Consumption on Model Simulation Properties
* Rational Expectations in Macroeconomic Models

Session 4: Practical Session

* A Policy Game Using the Fair Model of the US Economy

http://www.timberlake.co.uk/training...wsNYapril.html