Syllabus for Financial Econometrics

Lecturer: S.Gelman
Syllabus for Financial Econometrics
Course description:
Financial Econometrics course is a one semester course for the 1 st year MSc ICEF students. It is
designed to cover some essential tools for working with financial data, including forecasting
returns, volatility and event studies. We focus on the empirical techniques which are mostly used
in the analysis of financial markets and how they are applied to real-world data.
Teaching Objectives:
On completion of the course students should be:
• Familiar with the basic tools available to financial economists for testing theories,
estimating the parameters of economic relationships in financial markets and forecasting
financial variables.
• Able to read, understand and replicate the results with the use of real-world data of some
core papers in finance using standard computer packages.
• Trained in the writing up of reports and their subsequent communication, both in written
form and in the context of a presentation to a class.
Prerequisites
The students are expected to have a thorough mathematical and statistical background. Prior
mathematical knowledge should include multivariate calculus, linear algebra and matrix analysis
(in particular, basic rules of matrix differentiation). As for the statistical background, students
should be familiar with key concepts of probability theory as well as with hypothesis testing,
linear regression, maximum likelihood and basics of time-series analysis. However, the most
important prerequisite is a solid background in the key concepts of finance theory, in particular:
risk-aversion and expected–utility theory, static mean-variance portfolio theory, CAPM and APT
as well as dynamic asset pricing models.
The demands of the course are likely to be computation-intensive therefore some rudimentary
programming and data analysis skills are necessary.
Teaching methods
The following methods and forms of study are used in the course:
- lectures (2 hours a week)
- tutorials (2 hours a week, half of the tutorials is devoted to theoretical and applied analysis,
and another half is conducted in the computer room and is devoted to practical applications
of the econometric methods studied in the course)
- home assignments for each topic consisting of theoretical and applied parts (several
assignments, not exceeding 7 in number)
- presentation and discussion of contemporary research in empirical finance related to the
topics under study (4-6 hours on total, adequately distributed among the topics; presentations
take place during lecture or class hours)
- teachers’ consultations
The following method can be used in the course:
- Research paper (essay). Students are expected to come up with research ideas on the topics
under study and apply techniques learned in the course. Alternatively they can choose a
research question from a list. The essay should be a 10-12 pages (3000-4000 words)
empirical study with the use of real-world financial data (from Datastream, Bloomberg,
economic data bases, ICEF information system, etc.). The estimation should be conducted in

a computer room, using Econometric Views, RATS and other available statistical
applications.
In total the course includes: 34 hours of lectures and 34 hours of tutorials.
Textbooks, journal articles and other literature used in the course.
Students should use the following two books as the main course text:
1. John Campbell, Andrew Lo, Archie MacKinlay (1997). The Econometrics of Financial
Markets, Princeton University Press. (CLM)
2. Ruey S. Tsay (2002). Analysis of Financial Time Series. Wiley. (RT)
By several topics it is strongly recommended to refer further to the following textbooks:
3. Walter Enders (2003). Applied econometric time series, Wiley. (WE)
4. Juergen Franke, Wolfgang Haerdle, Christian Hafner (2004). Statistics of Financial
Markets, Springer. (FHH)
5. Chris Brooks (2002). Introductory econometrics for finance, Cambridge University Press.
(CB)
6. John H. Cochrane (2005). Asset Pricing, Princeton University Press. (JC)
7. Hamilton, J. (1994), Time Series Analysis, Princeton University Press, Princeton. (JH)
More precise references and further reading (e.g. journal articles) are provided in course outline
below, following the corresponding topic.
Course outline
1. Properties of financial data
CLM: Ch. 1
RT: Ch. 1
The main properties of financial data will be discussed in this introductory section.
First, I will address the sources of getting the data. We are going to discuss such databases as
Bloomberg, Datastream and CRSP-Compustat, as well as some open sources available on the
Internet. We shall cover basic database usage and special features, necessary transformations
of raw financial data for meaningful analysis.
Second, we will proceed with main statistic properties of financial data: stationarity issue,
distribution functions and so on.
(4 lecture hours; 4 class hours)
2. Forecasting and return predictability
2.1. Quick Review of Time Series Models and Forecasting
RT: Ch. 2
WE: Ch. 2
JH: Ch. 4
(8 lecture hours; 8 class hours)

2.2. Tests of return predictability
CLM: Ch. 2-3
Lo, A., 1991, "Long-Term Memory in Stock Market Prices," Econometrica 59,
1279-1313.
Lo, A. and C. MacKinlay, 1988, "Stock Market Prices Do Not Follow Random
Walks: Evidence from a Simple Specification Test," Review of Financial Studies
1, 41-66.
Bossaerts, P., and P. Hillion, 1999, Implementing Statistical Criteria to Select
Return Forecasting Models: What Do We Learn? Review of Financial Studies 12,
405-428.
Fama, E. and K. French, 1988, Dividend Yields and Expected Stock Returns,
Journal of Financial Economics 22, 3-26.
(2 lecture hours; 2 class hours)
2.3. Forecast Evaluation
Diebold, F. X. and Lopez, J. A.: 1996, Forecast evaluation and combination, in G.
Maddala and C. Rao (eds), The Handbook of Statistics, Vol. 14, Elsevier North
Holland.
Sullivan, R., Timmermann, A. and White, H.: 1999, Data-snooping, technical
trading rule performance, and the bootstrap, Journal of Finance 54, 1647–1691.
Patton, A. and Timmermann, A.: 2005, Properties of optimal forecasts under
asymmetric loss and nonlinearity, forthcoming in Journal of Econometrics.
White, H.: 2000, A reality check for data snooping, Econometrica 68, 1097–1126.
(2 lecture hours; 2 class hours)
3. Volatility
RT: Ch. 3
WE: Ch. 3
3.1. GARCH
Engle, R. F.: 1982, Autoregressive conditional heteroscedasticity with estimates
of the variance of United Kingdom inflation, Econometrica 50, 987–1008.
Bollerslev, T.: 1986, Generalized autoregressive conditional heteroskedasticity,
Journal of Econometrics 31, 307–327.
Engle, R. F., Lilien, D. M. and Robins, R. P.: 1987, Estimating time varying risk
premia in the term structure: The arch-m model, Econometrica 55, 391–407.
Engle, R.F., Patton, A., 2001, What good is a volatility model?, Quantative
Finance 1, 237-245
(6 lecture hours; 6 class hours)
3.2. Asymmetric GARCH
Glosten, L. R., Jaganathan R., and Runkle D. E. (1993). On the relation between
the expected value and the volatility of nominal excess return on stocks. Journal
of Finance 48, 1779-1801.
Zakoian, J. M. (1994). Threshold heteroscedastic models. Journal of Economic
Dynamics and Control 18, 931-955
Nelson, D. B. (1991). Conditinal heteroscedasticity in asset returns: A new
approach. Econometrica 59, 347-370.
(4 lecture hours; 4 class hours)
3.3. Volatility specification checking
Wooldridge, J. M.: 1990, A unified approach to robust, regression-based
specification tests, Econometric Theory 6, 17–43.
Andersen, T. G. and Bollerslev, T.: 1998, Answering the skeptics: Yes, standard
volatility models do provide accurate forecasts, International Economic Review
39(4), 885–905.

Lunde, A. and Hansen, P. R.: 2001, A forecast comparison of volatility models:
Does anything beat a garch(1,1)?, Working Papers 2001-04, Brown University,
Department of Economics.
(2 lecture hours; 2 class hours)
4. Event Study Methodology
CLM: Ch. 4
Boehmer, E., Musumeci, J. and A. Poulsen, 1991, Event-Study Methodology under
Conditions of Event-Induced Variance, Journal of Financial Economics 30, 253-272.
Fama, E., Fisher, L., Jensen, M. and R. Roll, 1969, The Adjustment of Stock Prices to New
Information, International Economic Review 10, 1-21.
Prabhala, N., 1997, Conditional Methods in Event Studies and an Equilibrium Justification
for Standard Event-Study Procedures, Review of Financial Studies 10, 1-38.
(2 lecture hours; 2 class hours)
Typical problems and assignments for exams and coursework
I. Typical empirical class assignment (Topic: 4. Volatility):
The file “m-gmsp5099.dat” contains monthly log returns, in percentages, of General Motors
stock and S&P 500 index from 1950 to 1999.
(a) Build a Gaussian GARCH model for the monthly log returns of S&P 500 index. Check
the model carefully.
(b) Is there a Summer effect on the volatility of the index return? Use the GARCH model
built in part (a) to answer this question.
(c) Are lagged returns of GM stock useful in modeling the index volatility? Again, use the
GARCH model of part (a) as a baseline model for comparison.
II. Typical theoretical class assignment (Topic 4):
Derive multistep ahead forecast for a GARCH (1, 2) model at the forecast origin h.
III. Typical research paper (essay) assignments:
1. Seasonalities in Lukoil daily stock returns (Topic 3)
Grade determination
About 80% of the final grade is determined by the exam paper at the end of the course and
home assignments can determine up to 20% of the final grade.
Students should be able to use computers and econometric software to solve empirical exam
problems. About 50% of the exam problems is of empirical nature. Taking the example
assignments above, students should have about 30 minutes for an empirical assignment such
as I. and about 10 minutes for a theoretical assignment similar to II.
One could introduce an extended learning achievement evaluation scheme, and include a
research paper (see III.) as well as a midterm test, which could receive weights of about 15%
each at the expense of the final exam’s weight.