Methods for Empirical Asset Pricing with Large Data Sets

The SoFiE Financial Econometrics Schools are annual week-long research-based courses for Ph.D. students and new faculty in financial econometrics. For the first two years, the Summer School was held at Oxford University’s Oxford-Man Institute and in 2014 it moved to Harvard University. In 2015 and 2016, it was held in Brussels. Since 2017, The SoFiE Financial Econometrics Summer School takes place in North America, Asia and Europe. The 2020 European edition will continue to take place in Bruxelles, at the National Bank of Belgium and is co-organised by the National Bank of Belgium, the Free Universtiy of Brussels, the KULeuven, Ghent University and the UCLouvain.

The editorial board for these annual series is made up of the following professors: Torben G. Andersen(Northwestern University), Luc Bauwens (Catholic University of Louvain, Francis X. Diebold (University of Pennsylvania, past President of SoFiE), Eric Ghysels (University of North Carolina, Chapel Hill, Founding Co-President of SoFiE), Tim Bollerslev (Duke University and President SoFiE), Eric Renault (Brown University and past SoFiE President), Neil Shephard (Harvard University), Viktor Todorov (Northwestern University)

Local organizing committee:

Luc Bauwens (UCLouvain), Kris Boudt (VUB, UGhent), Geert Dhaene (KU Leuven), Leonardo Iania (UCL), Raf Wouters (NBB).

Lecturers:

Olivier Scaillet

is Professor of Finance and Statistics at the University of Geneva, senior chair at the Swiss Finance Institute, and deputy director of the Geneva Finance Research Institute. Olivier Scailled holds a Ph.D. from the Université Paris IX Dauphine. His research interest and expertise include several areas of finance and econometrics such as term structure modelling, risk management in finance and insurance, derivative pricing, asset allocation and specification tests. His research output appeared in top tier econometrics and finance journal such as Econometrica, Journal of Financial Economics, Journal of Econometrics and Econometrics Theory. For the academic year 2019-2020 he holds the Chair International Francqui Professor.

Fabio Trojani

is Professor of Finance and Statistics at the University of Geneva and Senior Chair at Swiss Finance Institute. He holds a Ph.D. in econometrics and finance from the University of Zurich. His research interest and expertise include several areas of finance and econometrics such as asset pricing, financial econometrics, portfolio theory, term structure modelling, option pricing, and credit risk. His research output appeared in top tier statistics, econometrics and finance journal such as Journal of Finance, Review of Financial Studies, Journal of Econometrics and Journal of the American Statistics Association. He is co-editor of the Journal of Financial Econometrics.

Course Description

This summer school covers various selected topics from particularly active recent research areas in asset pricing and financial econometrics, with an emphasis on techniques useful to analyze large dimensional datasets. The course starts with a self-contained approach for the analysis of asset markets, based on generalized notions of nonparametric stochastic discount factors (SDFs) that may incorporate frictions and/or nonzero pricing errors. Applications of the methodology to several empirical asset pricing questions are presented, such as model-free SDF recovery problems, tests of asset pricing models with regularized SDFs, or studies of linear pricing rules compatible with the Arbitrage Pricing Theory. The Arbitrage Pricing Theory is the foundation for the empirical study of asset risk premia with linear factor models. In this context, the course introduces a suitable asymptotic framework to study two-pass cross-sectional estimators of conditional factor risk premia in unbalanced panels, where the cross-sectional and time series dimensions of the panel are allowed to grow simultaneously. The setting naturally accommodates also goodness-of-fit tests for conditional linear factor structures. The methodology is applied to study the properties of risk premium dynamics in large panels of stocks. The course finally covers multiple testing problems often found in applied econometrics, where many null hypotheses may have to be tested simultaneously. In such settings, one would ideally like to reject all hypotheses that are false. However, this cannot be achieved with certainty, given a finite amount of data. Hence, multiple testing methods aim to target as many true rejections as possible, while avoiding too many false rejections. We present various statistical solutions to these problems and illustrate their application to the generation of alpha by mutual funds and hedge funds, the performance analysis of trading strategies, and the detection of jumps in high-frequency data.

Applications

Applications should be sent to leonardo.iania@uclouvain.be (with the words "SoFiE School 2020" in the subject box). The applications should include a full CV and motivation letter of half a page explaining why attending this course would be helpful to the applicant's research work. The application deadline is 30 April 2020. Decisions will be emailed out by 07 May 2020.

Paper presentations:

The course will offer to a limited and selected number of course participants an opportunity to present their current research and receive feedback from the instructors and other course participants. Students interested in making a presentation (which is entirely optional) should indicate so on their application and submit the research paper that will form the basis of their presentation. Students who are selected to make a presentation will be informed at the same time as they receive their admission decisions.

Fee for attending the school:

250 Euro for academics and 750 Euro for non-academics. All accepted participants are expected to be members of the Society for Financial Econometrics or to join before their place is confirmed. Further info on how to join SoFiE is available at http://sofie.stern.nyu.edu/membership

(where a student membership option is available).

Tentative schedule

Monday:

13:00: Registration

14:00-17:00: Pricing errors, Smart Stochastic Discount Factors (S-SDFs) and markets frictions, Arbitrage Free Foundation of Smart S-SDFs, APT- Consistent S--SDFs, good deal bounds and regularization devices.

Tuesday:

9:00-12:00: Minimum Dispersion S-SDFs and S-SDF bounds, Duality results and characterization of minimum dispersion S-SDFs, Estimation and inference for S-SDFs and S-SDF bounds.

12:00-14:00: Lunch

14:00-16:00: 3 selected presentations

Wednesday:

9:00-12:00: APT consistent S-SDF pricing (applications), International market segmentation and global factor structures of exchange rates (applications). Multiple Testing Procedures.

12:00-14:00: Lunch

14:00-16:00: 3 selected presentations

Thursday:

9:00-12:00: False Discoveries in Mutual Fund Performance: Measuring Luck in Estimated Alphas. Detecting spurious jumps in high frequency data.

12:00-14:00: Lunch

14:00-16:00: 3 selected presentations

Friday:

9:00-12:00: Technical Trading Revisited: False Discoveries, Persistence Tests, and Transaction Costs. Factor Modelling in Finance. Sponsors: Center for Operations Research and Econometrics (CORE), FINS@VUB, Foundation Louvain, KU Leuven, Louvain Finance, National Bank of Belgium, TreeTop AM, UGhent.