Time Series Econometrics and Forecasting
This core module provides an in-depth understanding of the fundamental concepts of time series econometrics and forecasting, together with the practical skills to use econometric software to model and forecast economic time series and identify the models with the best forecasting abilities. It builds on the Foundations of Econometrics core module and prepares students for MSc and PhD research.
Prerequisites
This module builds on Foundations of Econometrics, which should be completed first. Applicants require an upper second-class degree in economics, econometrics or a related subject, and may be asked for a recognised English language qualification (IELTS Academic 6.5 overall, with a minimum of 6.0 in each element).
Delivery
Four hours of lectures per day across five days, plus a one-hour tutorial each day. All sessions are live online at a scheduled time and recorded for review.
Study Hours
The module carries 20 credits: approximately 200 hours of study, of which 25 are contact hours and 175 are private study.
Software Used
EViews 14
Credits
20 CATS credits
Programme Faculty
Programme Faculty
Programme Faculty
Programme Faculty
Key Outcomes
- —Understand how to model a univariate or multivariate time series process
- —Understand the maximum likelihood estimation method
- —Distinguish between stationary and nonstationary series and understand the implications of using nonstationary series
- —Understand how to forecast a univariate or a multivariate time series process
- —Be able to conduct forecast evaluation with statistical and economic loss functions
- —Build, estimate and forecast from univariate and multivariate time series models using econometric software
- —Understand and critically evaluate recent research in forecasting
Desired Skills
- —Engage in abstract thinking by extracting the essential features of complex systems to facilitate problem solving and decision-making
- —Communicate and present complex arguments in oral and written form with clarity and succinctness
- —Present, interpret and analyse information in numerical form
- —Utilise statistical and other packages effectively
- —Apply econometric techniques to aid understanding of the financial and macroeconomic environment
- —Work effectively both individually and within a team environment
Lecture Programme
- WK 01
Lecture 1: Introduction and key concepts
- ·Definitions of time series
- ·Strictly and weakly stationary processes
- WK 02
Lecture 2: Moments, stochastic processes and lag operators
- ·Conditional moments
- ·Simple stochastic processes
- ·Lag operators and data transformations
- WK 03
Lecture 3: The autoregressive model of order 1
- ·Definition of AR(1)
- ·Stability and moments
- WK 04
Lecture 4: ARMA models
- ·Overview of ARMA models
- ·Stability and moments
- WK 05
Lecture 5: ARMAX models
- ·Overview of ARMAX/ADL models
- ·Stability and moments
- ·Short-run and long-run multipliers
- WK 06
Lecture 6: Forecasting and impulse response functions
- ·Forecasting and IRFs for ARMA models
- ·Forecasting and IRFs for ARMAX plus ARMA models
- ·Examples
- WK 07
Lecture 7: ARMA model estimation
- ·The maximum likelihood estimator
- ·Estimator properties
- WK 08
Lecture 8: ARMA model specification
- ·General-to-specific estimation
- ·Residual testing
- WK 09
Lecture 9: Testing and validation
- ·Testing and diagnostic checking
- ·ARMA model validation and comparison
- WK 10
Lecture 10: Multivariate time series
- ·Concepts and properties
- ·VARs and seemingly unrelated regression
- WK 11
Lecture 11: Estimation and forecasting of VAR models
- ·Estimation by ML
- ·Forecasting vector models
- ·Impulse response functions
- WK 12
Lecture 12: Unit roots and stationarity tests
- ·Introduction to the notion of stationarity and unit roots
- ·The Dickey–Fuller unit root test
- ·Other stationarity and unit root tests
- WK 13
Lecture 13: SARIMA models
- ·The Box–Jenkins methodology
- ·Integrated ARMA models
- ·Seasonal ARIMA models
Assessment
Online coursework submission with the support of your module lecturer and tutor. All sessions are conducted live online at a scheduled time and recorded for review.