Intermediate Econometrics - 5QQMN938

Dr Jack Fosten

5QQMN938 Intermediate Econometrics

Module Outline 2022/23
This course is an intermediate econometrics module which will focus on the
models required to analyse time series data. Students will gain a deep
understanding of several types of time series modelling approaches. This
will enable them to make real-world forecasts of important economic and
financial series, useful for further study and careers in economics, finance,
retail and others.

Contact Details
Lecturer: Dr Jack Fosten (he/him), Senior Lecturer in Economics
E-mail: [email protected]
Office: Bush House (N)4.17
Office hours: Mon 11.00-12.00, Tues 14.00-15.00, (please book slot online)

Assessment
Forecasting Project (40%) – Assignment to be submitted on Keats will be released Tuesday 14th
February 2023 at 6pm, due Monday 20th March 2023 at 10am.
Final Exam (60%) – This will be an open-book format with a 24-hour window. The release date
will be made available during the semester. Guidance on the format of this assessment will be
provided on the Keats page.

Pre-Requisites
Students must have taken the pre-requisite module 5SSMN932 Introduction to Econometrics.
Students on BSc Economics & Management and BSc Business Management will not be allowed
to take this module unless they have the pre-requisites. This course is compulsory for BSc
Economics students

Module Structure
The course will run for ten weeks and will follow a hybrid model of pre-recorded online videos
along with a weekly in-person workshop and small-group tutorials:
Pre-recorded material (online): There will be between one and two hours of online lecture
material per week, including short videos and activities covering the core theoretical material for
each topic. The pre-recorded format is ideal for the mathematical and statistical content of the
module where students can pause and rewind over technical material and use activities to
reinforce their knowledge. This is based on my own past experience of running the module, from
excellent student feedback, and from the pedagogic theory of delivering quantitative methods.
Workshop (in-person): We will all meet for a two hour per large-group workshop. The
workshops are designed to 1) provide a brief recap and reinforcement of the theory from the
online material, 2) as a chance for Q&A and interaction and 3) to apply the theory of the course to
real-world data examples. Students will also be given a chance to submit their own real-world
applications using interesting data of their choosing which can be presented in the workshop. This
allows students to co-create the content of the module.
Tutorials (in-person): There will be one hour of small-group tutorials per week based on
computer problems issued in a weekly problem set. This will allow students hands-on practice of
using Stata to obtain and interpret econometric results. In the problem sets, students will also be
given mathematical problems which will be introduced in the lecture activities to practice in their
own time. For these mathematical parts, written solutions and interactive video solutions (made
before the transition to blended learning, apologies for the double videos!) will be provided.

Real-World Examples
This course will be grounded in important real-world applications. Examples to include: forecasting
inflation, employment, real GDP growth, house prices, hospital attendances and others. We will
take a global perspective and look at data from various countries to make the course inclusive to
our diverse set of students. We will pay attention to the changes in economic time series resulting
from the COVID crisis.

Module Objectives
This in an intermediate econometrics course focussing on the methods to analyse time series
data. The objectives of the module are:
• To demonstrate an understanding of theoretical and practical concepts in time series
econometrics and solve mathematical problems;
• To build econometric models which are used for forecasting and policymaking, and
assess their performance and limitations;
• To apply econometric models to real-world data, and to express the meaning of
econometric results to a wider audience;
• To build practical and transferable computing skills using the programme Stata;
• To demonstrate pre-requisite knowledge necessary for further study in final-year modules
(such as Advanced Econometrics) and MSc programmes at the highest-ranked
institutions.

Employability Skills
This course will give students very relevant quantitative skills for further study and careers in
economics, policymaking, consultancy, banking and finance, retail and others. Specifically:
• Problem-solving and analytical skills;
• Data analysis and data preparation skills;
• Econometric model-building and forecasting;

Page 2 of 4
 • Creative presentation of technical results;
• Confidence and ability in econometric software, including elements of code-writing

Reading
The core reading for this course will be the lecture notes and tutorial problems. As this is an
intermediate course in time series econometrics, we will take the topics typically found in
introductory econometrics textbooks but develop a deeper theoretical and practical understanding
of the models. The core introductory reading material for this course will be Part 4 of the textbook:

Stock, J. and Watson, M. “Introduction to Econometrics” 3rd Ed., Pearson.

This book gives a good introduction to the themes in this course. Through the lectures and
tutorials, students will gain a much deeper understanding of the topics than in Stock and Watson.
They are therefore encouraged to study the lectures and tutorials and supplement this reading
with the more advanced textbook:

Enders, W. “Applied Econometric Time Series” 4th Ed., Wiley

Another useful textbook with coverage of time series methods is:

Verbeek, M. “A Guide to Modern Econometrics” 5th Ed., Wiley

Lecture Topics
1. Introduction to Time Series
Time series data, properties, transformations, outlier detection, autocorrelation, stationarity, simple
time series regression
Stock and Watson (S&W): 14.1 and 14.2
2. Univariate Models – Autoregression Part 1
Intro to the AR(1) model, properties (mean, variance, ACF), estimation by OLS, forecasting
S&W: 14.3, 14.4 and Appendix (also Enders: 2.3, 2.7)
3. Univariate Models – Autoregression Part 2
Intro to the AR(p) model, lag operator notation, stationarity conditions, lag length selection (AIC,
BIC), forecasting
S&W: 14.5 (also Enders: 2.3, 2.4, 2.7)
4. Univariate Models - MA and ARMA
Intro to MA(q) and ARMA(p,q) models, properties of MA(1) model, estimation by maximum
likelihood, forecasting
S&W: Appendix 14.4, (also Enders: 2.2, 2.4, 2.5)
5. Multivariate Models
Intro to ADL models, dynamic causal effects, cumulative multipliers, HAC standard errors,
Granger causality, forecasting, nowcasting

Page 3 of 4
S&W: 14.4, 15.1 – 15.4 (also Enders: 5.2)
6. Multi-step Forecasting and Pseudo Out-of-Sample Methods
Iterated multi-step forecasts in univariate and multivariate models, vector autoregression (VAR),
direct multi-step forecasts, fan charts, pseudo out-of-sample methods
S&W: 16.1, 16.2, 14.7 (also Enders: 2.9, 5.5, 5.6)
7. Nonstationarity – Trends and Unit Roots
Deterministic and stochastic trends, random walks, unit roots, Dickey-Fuller unit root testing
S&W: 14.6, 16.3 (also Enders: 4)
8. Nonstationarity - Spurious Regression & Cointegration
Intro to spurious regression, order of integration, cointegration, Engle-Granger cointegration
testing, error correction models
S&W: 16.3, 16.4 (also Enders: 6)
9. Nonstationarity – Seasonality and Breaks
Seasonal adjustment, seasonal AR models, seasonal random walks, structural break testing
(Chow and Quandt) and estimating the break date
S&W: 14.7 (also Enders: 2.11, 2.12)
10. Volatility Models
Intro to volatility, the ARCH and GARCH models, estimation and forecasting
S&W: 16.5 (also Enders: 3)

Inclusive Module Values

This module is intended to be an inclusive module which is flexible, equitable and transparent for
all. In some of our examples we will look at global perspectives using data from across the world,
and we will also consider important topics like the gender employment gap. Students will also be
invited to share their own data and experiences to promote inclusivity on the module. If you have
any ways in which you feel the module inclusivity can be improved, please email
[email protected].

Page 4 of 4