Financial Econometrics

Course Description

The course has two objectives. The first is to introduce students to the application of statistical and econometric techniques to analyze financial data. The methodological part of the course will cover three main areas:

1. Linear regression model and its application to understand the factors driving stock returns and to measure their riskiness
2. Time series models use the past of a variable to forecast its future; one of the applications that we will consider is to forecast quarterly revenue of a company
3. Volatility models are time series model that are used to forecast the variance or standard deviation of a financial variable; we will consider the application of these models to financial risk management

The second goal of this course is to familiarize students with the R statistical and programming language. R is a widely used tool for data analytics that is open source, has a large base of users that contribute packages and discussions, and relatively easy to learn. Both my lecture notes and slides integrate the discussion of the methods and their application with the implementation in R in a learning-by-doing manner.

Learning Outcomes

At the end of the course students will be able to:

• Discuss the linear regression model and its assumptions and apply it to financial problems
• Build time series models for forecasting economic and financial variables
• Measure financial risks using different distributional assumptions
• Develop, implement, and present in written and oral forms a financial data analysis project
• Proficiently use the R statistical language

Required resources

• S. Manzan, Introduction to Financial Econometrics pdf html
• Class slides
• Datacamp
  • Datacamp is a website that offers courses for aspiring data scientists. It is free for academic use and I created a group where assignments will be posted
    • Each assignment will consists of a course (including several units) or a few chapters from a course; each unit includes a video that explains a task in R followed by exercises that are completed in the browser
    • I will send an invitation to join the group to the email address provided in Blackboard; accept the invitation and you will be asked to create an account or login with your existing account credentials
    • You will not need to complete a course again if you already did it
    • Weekly assignments that count toward your final grade
• WRDS Wharton Research Data Services
  • WRDS is an online platform to access financial data that is used in academia and industry
  • I created a class that allows you to access the datasets subscribed by the College
  • The class username and password to access WRDS are available in Blackboard and they expire at the end of the semester
  • NB: you agree not to be share these credentials with any other individual outside of the class
  • The account can accommodate up to 15 simultaneous users
  • Most empirical assignments will be based on WRDS datasets

Additional resources

[Where I indicate free below means that you can access or download the book for free if you are in campus or access the library remotely with your Baruch credentials]

• Tsay, An Introduction to Analysis of Financial Data with R, Wiley (free)
• Tsay, Analysis of Financial Time Series, Wiley
• Christoffersen, Elements of Financial Risk Management, 2nd Edition, Academic Press (free)
• R Core Team, An Introduction to R
• Zuur, Ieno, Mesters, A beginner’s Guide to R, Springer (free)
• Cowpertwait and Metcalfe, Introductory Time Series Analysis with R, Springer (free)
• Albert and Rizzo, R by Example, Springer (free)
• Wickham, ggplot2: Elegant Graphics for Data Analysis, Springer (free)
• Datacamp course on Intro to Computational Finance with R by Eric Zivot, UW
• Coursera course on R Programming (part of Data Specialization track)
• Swirl: a package to learn R in R
  • Install the package: install.packages("swirl")
  • Install a course: install_from_swirl("Course name") (details)
  • The courses related to our course are: (S1) R Programming, (S2) Data Analysis, (S3) Regression Models, (S4) Getting and Cleaning Data
• Datacamp tutorials: Functions, Importing Data in R part 1 part 2, Plotting, Apply, Loops, Data Frames, Excel Files, Data.Table (cheatsheet)

Software

We will use the R statistical language and the Rstudio environment

1. R download
2. Rstudio download (tutorial videos, cheatsheet)

Datacamp

Datacamp provides online courses to get started in data analytics in R and Python. A tentative schedule of assignments and topics is below. Notice that in some cases the assignment is a full Datacamp course that consists of 4-5 chapters, while in other cases only some chapters of a course are assigned. You should plan to spend approx 30 minutes per chapter, depending on the content.

A tentative schedule of assignments:

1. DC1: course Introduction to R
2. DC2:
   • chapter Compiling Reports from course Reporting with R Markdown
   • chapter Embedding Code from course Reporting with R Markdown
   • chapter Authoring R Markdown Reports from course Reporting with R Markdown
3. DC3: course Data Visualization with ggplot2 (Part 1)
4. DC4:
   • chapter Importing data from flat files with utils from course Importing Data in R (Part 1)
   • chapter readr & data.table from course Importing Data in R (Part 1)
   • chapter Importing Excel data from course Importing Data in R (Part 1)
5. DC5: course Data Manipulation in R with dplyr
6. DC6: course Intro to Statistics with R: Multiple Regression
7. DC? TBA

WRDS assignments

These assignments are designed to engage you with the techniques discussed in class using R. A typical exercise consists of obtaining data from WRDS or a public source (eg., Yahoo), estimate an econometric model, and interpret the results. The assignments will be done in Rstudio using the Rmarkdown package. Rmarkdown allows to write a document that incorporates text, R code and its output. The file is then compiled to Word without having to save the graphic files or copy and paste tables. The docx and Rmd files are then uploaded in BB by the due date.

Getting started with Rmarkdown:

• Website
• Get Started
• Cheatsheets

Final Project

The final project is the work of a group of at least 2 and no more than 4 students. The goal is to analyze a financial or economic problem and conduct an empirical investigation to answer a relevant question. The project should demonstrate the ability of the group to conduct a (relatively) complex analysis in R and to apply competently the statistical techniques discussed in class. The project idea could be inspired by a topic that you learned in this or another course, or that you read in a newspaper/ magazine/web/blog, as well as inspired by a real issue that you encountered at work. Try to think of a simple but relevant and interesting problem. Keep in mind that you might have a brilliant idea, but a necessary condition to finish an empirical project is data availability. So, check at an early stage that you have the data suitable to carry out the project (otherwise, switch to another topic).

The project is composed of a written report and an oral presentation. The Report should be max 10 pages long (font 12, spacing 1.5; submit separately the Rmd and Doc files) and structured as follows:

• Introduction: explanation of the topic of your project, its relevant, the main findings in the literature (has the literature reached a consensus on the topic or there are different views?), and, finally, a brief discussion of the main results.
• Survey of the literature: here you discuss the articles/papers that you found on the topic. It is probably a good idea to start with the earlier papers and explain in reasonable detail the technique, data, and results. The discussion of latest papers on the topic should be related to the earlier ones: what are the latest papers adding or doing different compared to the earliest ones? different data and/or econometric technique? are their results different?
• Model: provide an explanation of the empirical model that is used in the project. It can be a model suggested by economic theory or it could inspired by the problem at hand.
• Data: report the data source, the variables that are used, for how many individuals or for which period of time etc.
• Empirical Application: in this section discuss the results of the model estimation and the econometric issues that you addressed in arriving to the final specification. In the final specification you might have dropped some irrelevant variables, played with the functional form, allowed for heteroskedasticity, serial correlation etc.
• Conclusion: summarize the results of your project and how they are different or confirming earlier results.

The list below provides some paper that you might find useful as a starting point for your project:

• W.N. Goetzmann, L. Peng, J.Yen (2009), The Subprime Crisis and House Price Appreciation, NBER working paper 15334
• C. Himmelberg, C. Mayer, T. Sinai (2005), Assessing High House Prices: Bubbles, Fundamentals, and misperceptions, Staff Report n. 218, NY Fed
• C.S. Asness (2000), Stocks vs Bonds: Explaining the Equity Risk Premium, Financial Analysts Journal, 56(2), pg. 96-113
• C.J. Neely, D.E. Rapach, J. Tu, G. Zhou (2014), Forecasting the equity risk premium: the role of technical indicators, Management Science, 60(7), pg. 1772-1791
• R. Cantor and F. Packer (1996), Determinants and Impact of Sovereign Credit Ratings, FRBNY Economic Policy Review, October
• M. Scheicher (2008),How Has CDO Market Pricing Changed During the Turmoil? Evidence from CDS Index Tranches, ECB working paper, n. 910
• B. Barber, R. Lehavy, M. McNichols, B. Trueman (2003), Reassessing the Returns to Analysts’ Stock Recommendations, Financial Analysts Journal, 59(2), 88-96
• C. Baumeister and L. Kilian (2013),Forecasting the Real Price of Oil in a Changing World: A Forecast Combination Approach, CFS working paper
• C.B. Erb and C.R. Harvey (2006), The Strategic and Tactical Value of Commodity Futures, Financial Analysts Journal, 62(2), pg. 69-97
• A. Petajisto (2013), Active Share and Mutual Fund Performance, Financial Analysts Journal, 69(4), 73-93
• J. Hasanhodzic and A. Lo (2007), Can Hedge Fund Returns Be Replicated? The Linear Case, Journal of Investment Management, 5(2), pg. 5-45
• X. Lou and R. Sadka (2011), Liquidity Level or Liquidity Risk? Evidence from the Financial Crisis, Financial Analysts Journal, 67(3), pg. 51-62
• L.K.C. Chan, N. Jegadeesh, J. Lakonishok (1999), The Profitability of Momentum Strategies, Financial Analysts Journal, 55(6), pg. 80-90
• M.M. Copeland and T.E. Copeland (1999), Market Timing: Style and Size Rotation Using the VIX, Financial Analysts Journal, 55(2), pg. 73-81
• M. Haug and M. Hirschey, The January Effect, Financial Analysts Journal, 62(5), pg. 78-88
• S.C. Andrade, V. Chhaochharia, M.E. Fuerst (2013), Sell in May and go away … Just Won’t go Away, Financial Analyst Journal, 69(4), pg. 94-105
• M. Pojarliev and R.M. Levich (2008), Do professional currency managers beat the benchmark?, Financial Analysts Journal, 64, pg. 18-32
• L. Menkhoff, L. Sarno, M. Schmeling, A. Schrimpf (2012), Currency momentum strategies, Journal of Financial Economics, 106(3), pg. 660-684
• C.J. Neely and P.A. Weller (2011),Technical analysis in the foreign exchange market, St Louis Fed, working paper
• V. Bhansali (2008), Tail Risk Management, Journal of Portfolio Management, 34, pg. 68-75