Connect with IMF Institute
Internal Economics Training

Advanced Forecasting Methods (Self-Financed)

Deadline passed

Session No.: IT 19.10

Location: Washington, D.C., United States

Date: September 9-11, 2019 (3 days)

Primary Language: English

    Target Audience

    Officials, primarily in ministries of finance, economy, and planning, or in central banks.
    Back to top


    It is required that candidates have an advanced degree in economics, strong analytical skills, and a very good knowledge of English. Courses will be conducted in English with no interpretation.
    Back to top

    Course Description

    Professor Allan Timmermann (University of California, San Diego)

    The course introduces participants to a variety of advanced topics and approaches in economic forecasting. The first part of the course starts off by emphasizing the importance of grounding forecasts in the underlying economic objectives, as represented by the forecaster’s loss function. Often the difference between good and bad forecasting approaches hinges on how well they deal with changes to the underlying data generating process. The course therefore next addresses the consequences of model instability on forecasting performance and discusses strategies for dealing with such instability, using empirical illustrations from macroeconomics and finance.

    The second part of the course covers forecast evaluation, forecast comparisons, and monitoring of forecasting performance. Professor Timmermann will discuss which properties an optimal forecast should possess and introduce methods for evaluating whether a given forecast is efficient. He will also analyze the costs and benefits from using out-of-sample forecast evaluation methods—a common practice in economics. Ignoring the search across multiple models for a good forecasting model can introduce data mining biases, and this part will discuss ways to handle this problem. Next, Professor Timmermann will turn to methods for comparing the predictive accuracy of competing forecasts in situations with a pair of alternative forecasts as well as in settings with large numbers of forecasts. He will also analyze how real-time monitoring of competing models’ forecasting performance has the potential to improve upon the forecasting performance of individual models.

    Point forecasts do not reveal the uncertainty surrounding a given forecast and so many economic institutions have moved on to report interval and density (probability distribution) forecasts. The third part of the course introduces methods for generating these types of forecasts, covering both univariate and multivariate approaches. Finally, Professor Timmermann will discuss how to evaluate and potentially improve the accuracy of interval and density forecasts.

    Back to top

    Important Note for Internal Economics Training Courses

    Internal Economics Training (IT) courses are self-financed. The IMF will not charge officials for attending courses. However, all travel, insurance, hotel, and living costs will need to be covered by the agency sponsoring the participants.

    Back to top

    Apply Now - Application deadline approaching:

    IMF offers online training to government officials and public. Click Learn more to find our current offerings.

    Fiscal Policy Analysis (FPA)

    Arabic, September 8-19, 2019, Kuwait City, Kuwait

    Apply online by June 26, 2019

    Monetary Policy (MP)

    English, September 8-19, 2019, Kuwait City, Kuwait

    Apply online by June 28, 2019

    Financial Programming and Policies-Inflation Targeting (FPP-IT)

    Spanish, November 4-15, 2019, Washington, DC, United States

    Apply online by June 28, 2019

    Financial Sector Surveillance (FSS)

    English, September 23-27, 2019, Brasilia, Brazil

    Apply online by July 1, 2019