Econometrics Lunch
From GES
[edit] Econometrics Lunch
We meet Mondays, 12noon-1pm, McNeil 582 unless otherwise noted.
For a current schedule, and downloadable papers/slides when available, see http://boards.ssc.upenn.edu/ges/index.php/Econometrics_Lunch
[edit] Fall 2008: Schedule
- September 8
- Frank Diebold, "Real-Time Measurement of Business Conditions" (with S. Boragan Aruoba and Chiara Scotti; revised Sept.2008)
We construct a framework for measuring high-frequency economic activity using a variety of stock and flow data observed at mixed frequencies. Specifcally, we propose a dynamic factor model that permits exact filtering, and we explore the efficacy of our methods both in an empirical example and in a simulation study.
- September 15
- tba
- September 22
- Maxym Kryshko, "DSGE Model-Based Forecasting of Non-Modelled Variables" (with Frank Schorfheide and Keith Sill)
This paper develops and illustrates a simple method to generate DSGE model-based forecast for variables that do not explicitly appear in the model (non-core variables). We use auxiliary regressions that resemble measurement equations in a dynamic factor model to link the non-core variables to the state variables of the DSGE model. Predictions for the non-core variables are obtained by applying their measurement equations to DSGE model-generated forecasts of the state variables. Using a medium-scale New Keynesian DSGE model, we apply our approach to generate and evaluate recursive forecasts for PCE inflation, core PCE inflation, and the unemployment rate along with predictions for the seven variables that have been used to estimate the DSGE model.
- October 6
- Cristina Fuentes-Albero, "Financial Frictions and the Great Moderation"
[edit] Spring 2008: Schedule
- February 18
- Edith Liu, "Heterogenous Preferences and International Risk Sharing"
- February 25
- Aureo de Paula, "Interdependent Durations"
This paper studies the identification of a simultaneous equation model where the variable of interest is a duration measure. It proposes a game theoretic model in which durations are determined by strategic agents. In the absence of strategic motives, the model delivers a version of the generalized accelerated failure time model. In its most general form, the system resembles a classical simultaneous equation model in which endogenous variables interact with observable and unobservable exogenous components to characterize a certain economic environment. In this paper, the endogenous variables are the individually chosen equilibrium durations. Even though a unique solution to the game is not always attainable in this context, the structural elements of the economic system are shown to be semiparame- trically point identified. We also present a brief discussion of estimation ideas and a set of simulation studies on the model.
- March 3
- Gregor Baeurle, "Priors from DSGE Models for Dynamic Factor Analysis", [Slides]
- March 10
- Spring Break
- March 17
- No Meeting
- March 24
- Frank Schorfheide on Jungmo Yoon's paper: "Bayesian Conditional Density Estimation"
I develop a nonparametric Bayesian method for conditional density function estimation.
This method can be used to estimate the conditional density of a dependent variable given
covariates, and the transition density of the future observation given the past values.
Applications include the income distribution conditioned on individual specific
characteristics, and the distribution of returns conditioned on the state of the market.
To obtain the conditional density function estimate, I construct a posterior distribution
on the space of conditional quantile functions and then define a transformation from
conditional quantile function to the conditional density function. The posterior distribution
of conditional density function is approximated by multiple draws of conditional
density function, which in turn is obtained by drawing and transforming samples from the
posterior distribution of the conditional quantile function.
The posterior distribution of conditional quantile function is based on a likelihood
following the tradition of Jeffreysâs substitute likelihood and a Dirichlet process prior.
Posterior sampling can be conveniently done by Gibbs sampling. I establish the consistency
and the rate of convergence of my conditional density estimator. A simple data-driven
bandwidth selection rule is proposed. Finite sample performance of the Bayes estimator
is compared with the well-known double kernel method of Fan, Yao, and Tong (1996).
- April 7
- Karen Lewis, "International Equity Cross-Listings and Financial Integration" (with Gangadhar Darbha)
Foreign stock listing in the US has increased dramatically over the past decade, significantly reducing barriers to foreign investment by domestic residents. These declining barriers have led some to claim that optimal international diversification can be achieved using domestically traded stocks. At the same time, global capital markets appear to be more highly correlated. In this paper, we use the available history of foreign stock returns of companies that list in the United States to analyze whether their asset pricing relationships change over time. For this purpose, we use the structural time series break methodology of Bai and Perron (1998, 2001) to estimate the cross-sectional breaks in the asset pricing relationships of foreign stocks that list in the US. We then compare these structural asset pricing break estimates with cross-listing dates. While the literature has largely assumed that changes in asset pricing relationships have occurred before or during foreign listings, we find that most occur before stock cross-listings. We also analyze the asset pricing implications of the after cross-listing market, finding an overall increase in betas with respect to the world index after cross-listing.
- April 28
- No Meeting
- May 5
- Bernd Schlusche, "Data Snooping and Market Timing Rule Performance"
- May 12
- Leonardo Melosi and Cristina Fuentes-Albero, "Methods for Computing Marginal Data Densities from the Gibbs Sampler Output"
- May 19
- Aureo de Paula
