Fu Jen Catholic University Huei-Yu Chiu (n=Ø) Graduate School of Economics Office hours:
Spring 2007
Course number: G-6562-05918 Office tel:
Class hours: 9:10-12:00, Saturday Email: [email protected]
Special topics in Econometrics (l¾%Èçùæ)
Most macroeconomic and financial data are time series. This kind of data are collected according to time. The methodology of time series analysis is constructing models and forecasting based on the known past information. This course is designed to introduce the characteristic of time series models and make further forecasts. We attempt to cover time domain analysis and frequency domain analysis of time series.
For time domain analysis, stationary and nonstationary processes modelled with linear and nonlinear forms will be discussed respectively. Topics such as ARMA models, unit root tests, cointegration, error-correction model, threshold AR models, threshold cointegration and threshold ECM will be investigated. For frequency domain analysis, the concept of spectrum and periodogram will be studied. We will also introduce various bootstrap methods that simulate the distribution of interest via Monte Carlo approach.
Readings
1. Efron, B. and R. J. Tibshirani (1994), An Introduction to the Bootstrap, Chap- man & Hall/CRC.
2. Franses, P. H. and D. van Dijk (2000), Non-Linear Time Series Models in Empirical Finance, Cambridge University Press.
3. Fuller, W. (1995), Introduction to Statistical Time Series, Wiley-Interscience.
4. Hamilton, J. D. (1994), Time Series Analysis, Princeton University Press.
1
5. Kuan, C. -M. (2005), Basic Time Series Analysis, lecture notes.
6. Priestley, M. B. (2001), Spectral Analysis and Time Series, Academic Press.
Course Outline
Week 1 (3/3): Difference equations, lag operators, stochastic processes and stationarity
Week 2 (3/10): Stationary ARMA processes: model estimation and forecast Week 3 (3/17): Vector AR processes
Week 4 (3/24): Threshold AR models: SETAR, STAR, LSTAR models Week 5 (3/31): Markov switching model
Week 6 (4/7): Stationary processes in the frquency domain: spectral analysis Week 7 (4/14): Deterministic trend, stochastic trend & Wiener processes Week 8 (4/21): Unit root tests
Week 9 (4/28): Tests for stationarity
Week 10 (5/5): Cointegration: estimation & tests and error correction model Week 11 (5/12): Threshold cointegration and threshold error correction model Week 12 (5/19): Bootstrap methods: introduction
Week 13 (5/26): Wild bootstrap & pairs bootstrap Week 14 (6/2): Sieve bootstrap & block bootstrap
Week 15 (6/9): Bootstrap unit root test and cointegration test Week 16 (6/16): Final exam: term paper presentation
Grading
1. Homework (50%) 2. Term paper (50%)
2
