Latent Factor Analysis in Short Panels.
With the Lunch Seminar series, the Department of Finance is bringing eminent and up-and-coming researchers from around the world to Luxembourg.
Abstract:
We develop inferential tools for latent factor analysis in short panels. The pseudo maximum likelihood setting under a large cross-sectional dimension n and a fixed time series dimension T relies on a diagonal T ×T covariance matrix of the errors without imposing sphericity or Gaussianity. We outline the asymptotic distributions of the latent factor and error covariance estimates as well as of an asymptotically uniformly most powerful invariant (AUMPI) test based on the likelihood ratio statistic for tests of the number of factors. We derive the AUMPI characterization from inequalities ensuring the monotone likelihood ratio property for positive definite quadratic forms in normal variables. An empirical application to a large panel of monthly U.S. stock returns separates date after date systematic and idiosyncratic risks in short subperiods of bear vs. bull market based on the selected number of factors. We observe an uptrend in idiosyncratic volatility while the systematic risk explains a large part of the cross-sectional total variance in bear markets but is not driven by a single factor. Rank tests reveal that observed factors struggle spanning latent factors with a discrepancy between the dimensions of the two factor spaces decreasing over time.
About Patrick Gagliardini:
Patrick Gagliardini studied at the Polytechnical School in Zurich (ETHZ) where he graduated in Physics in 1998. In January 2003 he received a PhD from the Faculty of Economics of Università della Svizzera italiana (USI) for a thesis in Econometrics. In 2003 he has been a visiting fellow at the Laboratoire de Finance-Assurance of CREST (Paris) with a SNSF research grant. Between 2004 and 2006 he held an assistant professor position at the Faculty of Economics of the University of St. Gallen. Since 2012 he is full professor of Econometrics at USI. His research interests focus on econometrics and financial econometrics. He has published research papers on topics such as large panel factor models, nonparametric estimation, the Generalized Method of Moments in asset pricing, time series analysis, and credit risk. He teaches courses in Introductory Econometrics (Bachelor), Financial Econometrics (Master), Econometrics (PhD) and Time Series Analysis (PhD, together with prof. Fabio Trojani).
His research interest include:
1) Econometric methods: nonparametric methods, generalized method of moments (GMM), inverse problems, nonlinear time series, latent factor models, panel data
2) Financial econometrics: credit risk, asset pricing, option pricing
Patrick Gagliardini has competence in the following areas:
Big Data Dataset Disaggregated Data Econometrics Investment Decisions Risk Management Risk Premiums Systematic Risk
Language: English
This is a free seminar. Registration is mandatory.
The seminar will be held in person.
Cold lunches are provided to registered participants.
Supported by the Luxembourg National Research Fund (FNR) 17984041