Professors

Byeongseon Seo (Korea University)

Schedule

Tuesday
From 10:40
to 12:10
Thursday
From 10:40
to 12:10

Course description
This course is designed to provide a foundation of univariate and multivariate time series analysis for climate data. Topics include trend models, seasonality, cycle models, ARMA, VAR, GARCH, unit roots, cointegration, in-sample/out-of-sample forecasting, model selection, impulse-response analysis and variance decompositions, state space models, non-linear models, Bayesian approaches, and forecast evaluation.
Forecasting is one of the most professional areas with wide applicability in business and economics areas. These areas demand the experts who possess an understanding of forecasting, econometric tools to solve forecasting problems, and necessary computer skills. Foundation in economics, statistics, and mathematics is essential to meet these growing needs.
Class assignments will be passed out every two or three weeks approximately. These assignments will include both problem solving and empirical project. The assignments will be graded by the teaching assistant, and will be reviewed in the class. Class attendance and participation will be counted in the grade. Term project will be given for each group of three to four students. The goal of the project is to obtain hands-on experience in forecasting. Evaluation will be based on presentation and the term paper.
For the empirical work, econometric computer software will be studied. The recommended statistical software is one of Eviews, STATA, R. Tutorial session will be arranged to guide and assist you to this computer software.

Syllabus
1. Introduction to Forecasting
Overview
Linear Regression
2. Successful Forecasting (Chapter 2, 3)
Six Considerations
Statistical Graphics
3. Trend and Seasonality (Chapter 4, 5)
Modeling and Forecasting Trend
Modeling and Forecasting Seasonality
4. Cycles (Chapter 6, 7, 8, 9)
Characterizing Cycles
Modeling Cycles: MA, AR, and ARMA
Forecasting Cycles
Forecasting Model with Trend, Seasonal and Cyclical Components
5. Forecasting with Regression Models (Chapter 10)
Conditional Forecasting
Dynamic Models
Causality Tests
Midterm Exam
6. Evaluating and Combining Forecasts (Chapter 11)
Forecasting Accuracy
Forecast Encompassing
7. Unit Root and ARIMA Forecasting Models (Chapter 12)
Stochastic Trends
Unit Roots
8. Modeling Volatility (Chapter 13)
GARCH Process
Stock Market Volatility
9. Vector Autoregression (Chapter 10)
VAR Model Estimation
Impulse-Response
Variance Decomposition
10. Other Issues (Chapter 7, 12)
Regression with Integrated Variables
Cointegration: Model and Tests
Threshold Autoregressive Model
High Frequency Data Analysis
Spatial Data Analysis
Term Projects: Presentation
Final Exam

Evaluation method
Assignments: 20%
Midterm Exam: 30%
Final Exam: 30%
Term Project: 10%. Attendance: 10%

Bibliography
DIEBOLD, FRANCIS, X., (2007), Elements of Forecasting, 4th ed., Thomson South-Western.
CHEVALLIER, JULIEN, (2012), Econometric Analysis of Carbon Markets, Springer.
TSAY, RUEY, S., (2010), Analysis of Financial Time Series, 2nd ed., Wiley.

Venice
International
University

Isola di San Servolo
30133 Venice,
Italy

-
phone: +39 041 2719511
fax:+39 041 2719510
email: viu@univiu.org

VAT: 02928970272