B2219 - FORECASTING IN BUSINESS AND ECONOMICS

Learning outcomes

The aim of the course is to introduce the basic approaches to forecast economic variables with statistical models over short horizons, and with structural models over the long run. At the end of the course, students will be able to make their own model-based ex ante forecasts, and to evaluate the ex post forecasting ability of the professional forecasters activity.

Course contents

An introductory picture of forecasting in business and economics (FBE): governing many trade-offs

Bridge models as static/dynamic single-equation models: problems and advantages (GM, 1-3, 5

Midas models: relationships at mixed frequency of observations (GM, 12)

Factor models vs regressors’ selection algorithms: the curse of dimensionality (GM, 13)

Bayesian estimates to tackle the curse of dimensionality: from VAR to BVAR models (GM, 6-8)

Macroeconometric models and the amount of economic theory: RBC, DSGE and Hybrid-VECM

Forecast evaluation and combination for better predictions (GM, 4)

note: “GM, #” refers to single chapters of FBE course textbook, see below

Readings/Bibliography

Basic textbook:

Eric Ghysels and Massimiliano Marcellino, Applied Economic Forecasting using Time Series Methods, Oxford University Press, 2018 (referred to as GM).

Please note that chapters 1-3, 5, and 7 (excluding empirical examples) of GM must be read in advance (i.e., before the beginning of FBE lectures), as they review the main content of the Time Series Econometrics course from the first year.

Further readings will be provided during the lectures.

Teaching methods

During each of the 5 weeks of the FBE course, there will be 6 hours of traditional in-person teaching using a blackboard and slides.

Additionally, scheduled online meetings will be held every week to illustrate EViews applications and to assess your home assignments.

Assessment methods

A number of home assignments will be given during FBE. They must be completed weekly and will account for 50% of the grade. This approach ensures students are engaged throughout FBE and encourages active participation in all in-person lectures, as well as a number of online meetings.

In addition, there will be a final in-person exam, which accounts for the remaining 50% of FBE grade.

Students who do not attend all FBE lectures and online meetings are required to take a comprehensive written exam in person, covering all FBE topics.

Grades of FBE assessment: <18: not sufficient; 18-23: sufficient; 24-27: good; 28-30: very good; 30 e lode:outstanding.

Note for Erasmus (International) Students:

Since FBE is closely related to the content of the first-year "Time Series Econometrics" (TSE) course, selecting FBE without a full understanding of TSE may result in problems and extra work due to the integration of both courses.

Teaching tools

Online meetings and home assignments will utilize the EViews econometric software. The EViews 12 license is available in two ways:

(1) Purchase the University Edition online for $49.95, or

(2) Use the free EViews Student Version Lite.

Both options are illustrated at this link.

Alternatively, home assignments can be completed using R, but R online tutorials will be not provided.

Office hours

See the website of Roberto Golinelli