Home > Study > Summer schools > Methods Summer Programme > Quantitative Social Research Methods > ME411 Factor Models in Time Series with Applications in Macroeconomics and Finance
How to contact us

Tyrone Curtis
Programme Coordinator

Methods Summer Programme
London School of Economics
Houghton Street
London WC2A 2AE

Email: summer.methods@lse.ac.uk|
Tel: +44 (0)20 3199 5379

Register your interest here|

ME111: Factor Models in Time Series with Applications in Macroeconomics and Finance



Large datasets are becoming increasingly available to researchers and practitioners in many disciplines. In particular, during this “big data” revolution the analysis of high–dimensional time series has become one of the most active subjects of modern statistical methodology with applications in various areas of social science including finance, macroeconomics, and econometrics. Although the value of information is unquestionable, the possibility of extracting meaningful and useful information out of this large amount of data is also of great importance. To this end, several new analytical and computational techniques have been developed under the name of factor models.

The aim of this course is to give an introduction to factor models in time series analysis by teaching students the basic analytical methods and their applications to macroeconomics and finance via the use of Matlab software. These models are widely used in central banks for forecasting key macroeconomic indicators such as GDP and inflation. They are also used to study the impact of economic policies on economic activity and in validating models of the economy. Financial institutions adopt factor models for risk management.

This course is designed for postgraduates, academics and professionals with an interest in big data analysis and who have some analytical background in time series analysis.

Course benefits
After successful completion of this course, participants should be able to:

  • identify macroeconomic and/or financial policy problems that can benefit from factor analysis and consequently identify the appropriate dataset and methodology to be used
  • extract and analyse relevant information from large datasets
  • apply the analytical tools of time series analysis to the data using Matlab software
  • conduct empirical research in time series, i.e. to interpret the information extracted from the data in a critical way also in relation to the existing literature
  • forecast time series using many predictors

At least one semester of mathematical statistics with analytical treatment of estimation and inference, and at least one semester of multivariate calculus. Good background in methods of regression modelling and some basic familiarity with the analysis of multivariate time series.

The course consists of five daily lectures of three hours each, supported by four two-hour computer-based practical classes which will allow course participants to implement the lecture material in Matlab.

The course covers the following topics:

  • Motivations: the availability of large panels of time series and the value of information in macroeconomics, finance and other disciplines.
  • Exact and approximate factor models: the curse and blessing of dimensionality. We start by discussing principal component analysis as a useful dimension reduction technique for large panels of time series. This is the most simple example of factor model (the static model) which we then generalize to include all temporal relations among the considered variables (the dynamic model).
  • Estimation of factor models: we compare different models and discuss their estimation. The basic tools of multivariate time series analysis such as vector autoregressions and the Kalman filter will be introduced. 

We then apply these models to three main areas:

  • forecasting in real time of macroeconomics indicators such as Gross Domestic Product
  • policy analysis problems, i.e. the study of the dynamic reaction of observed variables to unexpected changes in policies such as monetary policies
  • optimization of financial portfolios.

All topics are of particular relevance for and widely used by researchers in central banks and national or international institutions. Examples based on real-data applications and taken from existing papers are presented and discussed during lectures and replicated during computer workshops.

Main Texts
There is no main textbook for this course. Lecture notes and all necessary material will be provided. 

Software Used
Matlab. No previous experience is expected.

Teaching Schedule
The following teaching schedule is indicative only, and is subject to change. Teaching takes place from 24-28 August 2015.

  Monday Tuesday Wednesday Thursday Friday
Introduction and principal component analysis Static and dynamic factor models Estimation and Kalman filter Macroeconomics applications Finance applications
Computer class Computer class Computer class Computer class Exam

All lectures take place from 10am-1pm. Computer practical classes take place in the afternoon.

A 2-hour final examination will take place on the afternoon of Friday 28 August 2015.

"Matteo, your course was absolutely mind-blowing. Thank you!"
Participant on the 2014 Methods Summer Programme

"I would like to thank Matteo for some very inspiring and enlightening lectures. They were some of the best I have ever experienced."
Participant on the 2014 Methods Summer Programme


Course details

24-28 August 2015

Lectures, practical classes

2-hour examination (optional)

New Academic Building, LSE

Teaching faculty
Dr Matteo Barigozzi
Department of Statistics

Tuition fees
Student rate: £935
Academic staff/charity rate: £1,250
Professional rate: £1,575


You may also be interested in...
Tools for Macroeconomists: The Essentials|
(17-21 August 2015)