EC2003 Financial Econometrics I

Course Structure

Lesson 1 - Introduction to Econometrics of Financial Markets

Lesson 2 - The Predictability of Asset Returns

Lesson 3 - The Capital Asset Pricing Model

Objectives

Develop the theoretical-practical fundaments of financial econometrics, teaching the techniques and tools for econometrics forecasting and estimation that are applicable to financial and economic models; the student will be capable of generating models, analysing results and infer from the econometric models and its financial applications, and criticising empirical work from academic and practitioners research.

Course Program

1. INTRODUCTION TO ECONOMETRICS OF FINANCIAL MARKETS (1 week)

(a) Prices, Returns. and Compounding

i. Definitions and Conventions

ii. The Marginal, Conditional, and Joint Distribution of Returns

(b) Market Efficiency

i. Efficient Markets and the Law of Iterated Expectations

ii. Is Market Efficiency Testable?

2. PREDICTABILITY OF ASSET RETURNS (3 weeks)

(a) The Random Walk Hypotheses

i. The Random Walk 1: IID Increments

ii. The Random Walk 2: Independent Increments

iii. The Random Walk 3: Uncorrelated Increments

(b) Tests of Random Walk 1: IID Increments

i. Traditional Statistical Tests

ii. Sequences and Reversals, and Runs

(c) Tests of Random Walk 2: Independent Increments

i. Filter Rules

ii. Technical Analysis

(d) Tests of Random Walk 3: Uncorrelated Increments

i. Autocorrelation Coefficients

ii. Portmanteau Statistics

iii. Variance Ratios

(e) Long-Horizon Returns

i. Problems with Long-Horizon Inferences

(f) Tests For Long-Range Dependence

i. Examples of Long-Range Dependence

ii. The Hurst-Mandelbrot Rescaled Range Statistic

(g) Unit Root Tests

(h) Recent Empirical Evidencei. Autocorrelations

ii. Variance Ratios

iii. Cross-Autocorrelations and Lead-Lag Relations

iv. Tests Using Long-Horizon Returns

3. INTRODUCTION TO THE CLASSICAL LINEAR REGRESSION MODEL (3 weeks)

(a) A regression model

(b) Regression versus correlation

(c) Simple regression

(d) The assumptions underlying the classical linear regression model

(e) Properties of the OLS estimator

(f) Precision and standard errors

(g) An introduction to statistical inference

(h) A special type of hypothesis test: the t-ratio

(i) An example of the use of a simple t-test to test a theory in finance: can US mutual funds beat the market?

(j) Can UK unit trust managers beat the market? (k) The overreaction hypothesis and the UK stock market (l) The exact signicance level (m) Estimation and hypothesis testing in practice { example: the CAPM

4. THE CAPITAL ASSET PRICING MODEL (2 weeks) (a) Review of the CAPM (b) Results from Efficient-Set Mathematics (c) Statistical Framework for Estimation and Testing i. Sharpe-Lintner Version ii. Black Version (d) Size of Tests (e) Power of Tests (f) Nonnormal and Non-IID Returns (g) Implementation of Tests i. Summary of Empirical Evidence ii. Illustrative Implementation iii. Unobservability of the Market Portfolio (h) Cross-Sectional Regressions

5. FURTHER ANALYSIS OF THE CLASSICAL LINEAR REGRESSION MODEL (2 weeks) (a) Generalising the simple model to multiple linear regression (b) The constant term (c) How are the parameters (the elements of the  vector) calculated in the generalised case? (d) Testing multiple hypotheses: the F-test (e) Multiple regression using an APT-style model (f) Data mining and the true size of the test (g) Goodness of t statistics (h) Hedonic pricing models (i) Tests of non-nested hypotheses

6. CLASSICAL LINEAR REGRESSION MODEL ASSUMPTIONS AND DIAGNOSTICS TESTS (4 weeks) (a) Statistical distributions for diagnostic tests 130 (b) Assumption 1: E(ut) = 0 (c) Assumption 2: var(ut) = sigma^2 < 1 (d) Assumption 3: cov(ui; uj) = 0 for i ̸= j (e) Assumption 4: the xt are non-stochastic (f) Assumption 5: the disturbances are normally distributed

(g) Multicollinearity (h) Adopting the wrong functional form (i) Omission of an important variable (j) Inclusion of an irrelevant variable

Evualuation Course evaluation will be done as following:

1. One case study (Goldman ITESM Investment Banking - Research Department) presented and evaluated by 9 written reports in English, and three presentations, that occur during the partial and final exam dates.

There will be 3 (three) evaluations during the semester. The final grade will be a weighted average of the three exams:

1. Semana i - From 26/09/2016 to 30/09/2016 - 5%

2. Partial Exam 1 - 12/09/2016 - 30% (Reports of: 15/08/2016, 29/08/2016 and 12/09/2016).

3. Partial Exam 2 - 24/10/2016 - 30% (Reports of: 19/09/2016, 10/10/2016 and 24/10/2016).

4. Final Exam - 28/11/2016 - 35% (Reports of: 31/10/2016, 14/11/2016 and 28/11/2016).

On each exam (Partial and Final), the groups will present jointly the reports (10 mins presentations in English), and this will have a weight of 30% of the reports for that exam, with the other 70% corresponding to the evaluation of written work. A report that is not presented on time will have zero grade.It will be considered approved the student with minimum assistance that obtains a total grade superior to 70 from the three evaluations. Will be considered failed the student with less than 70 points (69.5, for example). The student with more than 3 weeks of nonattendance will not be able to participate in the final exam.

References

[1] CAMPBELL, John; LO, Andrew; MACKINLAY, A. Craig. The Econometrics of Financial Markets. Princeton, New Jersey, Princeton University Press, 1997. [2] BROOKS, Chris. Introductory Econometrics for Finance. Cambridge, Cambridge University Press, 2008. [3] GREENE, William H. Econometric Analysis. Fifth Edition. New Jersey, Prentice Hall, 2003. [4] GUJARATI, Damodar N. Basic Econometrics. Fourth Edition. New York, McGraw-Hill, 2003.