Timberlake Consultants Ltd invite you to attend a four-day course in Central New York, on Energy Modeling and Forecasting. Several modern econometrics packages, including EViews and PcGive, will be used to demonstrate the use of the techniques under discussion.

The Course - This course will cover issues regarding the methodology, practice and implementation of econometric modelling with electrical time series data. Time series models in hourly, monthly and annual frequencies will be explored. Weather and temperature sensitivities will be examined. The use of time series, cointegration and error-correction techniques will be explored. In-class empirical presentations on the PC will demonstrate the approach and econometric questions as they arise.

The objective of the course is to serve as a refresher for economists and analysts on energy modeling and forecasting issues and introduce them to recent “developments” in the time series econometrics fields. The course is designed to be very applied. Hence, it is assumed that the participants have a basic knowledge of econometrics.

Model building is an attempt to characterize the properties of observed energy-economic data using simple parametric relationships which remain reasonably constant over time, account for the findings of previous models, and are interpretable in an energy-economic sense. The lectures will approach applied econometric issues from the LSE or Hendry tradition. The course focuses on the three key aspects of empirical model building: data properties, including integration and cointegration; dynamic specification, including the use of error correction models; and model evaluation and design. Electricity data from individual (utility) load regions in both developed and developing countries are analysed. A full agenda is also enclosed for your perusal.

EViews and PcGive will be the econometric software used in the course. No prior knowledge of the packages is required. Although offering an extensive GUI Windows interface, EViews offers also a batch language and programs will be provided to participants for production-oriented work. These programs can easily be converted to other time series languages such as TSP and Shazam.

Participants are invited to bring a data set for possible analysis during the course. These datasets should be sent to us, at least 1 week prior to the course, to give the course instructor enough time to familiarize with them and analyse the appropriateness of its use. Time will be set aside to provide “office hours” with the participants.

Who should attend - The course will be of particular interest to:

• New recruited energy modelers wanting a practical introduction to energy modeling
  
• More experienced energy modelers who want a refresher course or wish to learn new techniques
  
• Those who are required to interpret and present the results of models
  

The Course Leader - The course leader is Frederick L. Joutz, Associate Professor at the Department of Economics at George Washington University and Director of the Research Program on Forecasting.

Fred Joutz joined the Department in 1988. He received his Ph.D. in economics from the University of Washington in 1987. His areas of research interest are macroeconometrics, energy economics, time series, and forecasting. He contributes macroeconomic forecasts (known as the Benchmark Forecasts) to the Survey of Professional Forecasters (formerly the ASA/NBER Quarterly Outlook) produced by the Federal Reserve Bank of Philadelphia and the Ifo Institute for Economic Research . He is an associate editor for Energy Economics and the International Journal of Forecasting.

Cost - The cost of the course are:

1st Participant	$2800.00
2nd Participant +	$2500.00

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

Introduction

• Introduction to Course and Participants
  

A First Look at the Econometrics Software EViews: How it can be used?

• Exercise 1: Working with an Individual Series – summary statistics and plotting
  
• Exercise 2: Working with a Group of Series - summary statistics, tables and plotting
  
• Exercise 3: Inputting Data into EViews from different formats and with frequencies
  
• Exercise 4: Transformation and Creating Variables in EViews
  
• Exercise 5: Estimating a Regression and diagnostic testing
  
• Exercise 6: Forecasting in EViews
  
• Exercise 7: Building a Model and Forecasting
  
• Exercise 8: EViews and Report Writing

Hourly Load data

• Load Demand and Weather Sensitivity
  
• Hourly Load Curves
  
• Heating and Cooling Degree Day Measures
  
• A model of Hourly Demand
  
• Identification of Effects
  
• Forecasting

Day 2

Hourly Load data (continued)

• Combining Models of Hourly Demand
• Dynamic Pooled Time Series Models
• Forecasting with a System
  

Monthly Electricity Data (U.S. Aggregate, Individual Utility, Foreign)

• Sectoral Demand
• Seasonality
• Trends and Cycles
• Simple Time Series Models
  

Modeling and Forecasting Project – Introduction

Day 3

Monthly Electricity Data (continued)

• Energy Econometric Time Series Models
• Traditional Models
• Error Correction Models
• Forecasting Monthly Demand
  

Modeling and Forecasting Project - Presentations

Day 4

Annual Electricity Data (U.S. Aggregate and Regional Data, Taiwan)

• Sectoral Demand
• Trends and Cycles
• Simple Time Series Models
• Traditional Models
• Error Correction Models
• Forecasting Demand
  

Course Wrap-up

Top