Technological Forecasting & Decision Making

Learning Objectives

Upon completion of the course the students will be able to:
  1. Understand forecasting principles;
  2. Know how to apply statistical and artificial intelligent techniques to point and probabilistic forecasting;
  3. Know how to leverage high-performance computing techniques to improve forecasting system efficiency;
  4. Know how to document and present forecasting methodologies and results.

Text Book 
  • Rob J. Hyndman and George Athanasopoulos (2013) Forecasting: Principles and Practice, OTexts (available online:

Reference Books
  • J. S. Armstrong (2001), Principles of Forecasting: A Handbook for Researchers and Practitioners, Springer
  • William W. S. Wei (2005) Time Series Analysis: Univariate and Multivariate Methods, Addison Wesley
  • Trevor J. Hastie, Robert J. Tibshirani, & Jerome H. Friedman (2009). The elements of statistical learning: data mining, inference, and prediction. Springer.
Course Contents 
 1 Greetings; intro; terminology; reproducible research
 2 Exploratory data analysis
 3 Naive models; error analysis
 4 Time series analysis (exponential smoothing; ARIMA)
 Exam 1 In-class competition: forecasting with univariate models
 5 Regression analysis (multiple regression; logistic regression)
 6 L1 / support vector / Fuzzy / semi-parametric regression
 7 Artificial neural networks
 8 Clustering methods
 9 Forecast combination methods
 Exam 2 In-class competition: forecasting with explanatory variables
 10 Heuristic search; high-performance computing
 11 Hierarchical forecasting
 12 Probabilistic forecasting (conventional methods; Bayesian methods)
 13 Judgmental forecasting; emerging topics
 Exam 3 In-class competition: hierarchical probabilistic forecasting
 Project Phase 1. Literature review
 Phase 2. Reproducing results from a recent journal paper
 Phase 3. Forecasting NBA playoff results (or self-defined forecasting project approved by the instructor)

  • Homework: 6' x 5 = 30'
  • Exam: 10' x 3 = 30'
  • Project: 10' x 3 = 30'
  • Moodle discussion: 10'