« Back to Results
Marriott Philadelphia Downtown, Independence Ballroom II
Hosted By:
Econometric Society
Survey of Professional Forecasters, Blue Chip, or the Federal Open Market
Committee's Summary of Economic Projections. At a given point of time, these
surveys provide forecasts for macroeconomic variables at multiple
horizons. To track time-varying uncertainty in the associated forecast
errors, we derive a multiple-horizon specification of stochastic volatility.
Compared to constant-variance approaches, our stochastic-volatility
model improves the accuracy of uncertainty measures for survey forecasts.
structure. The conventional wisdom is to associate the common shocks with innovations to the level of
economic fundamentals. While second‐moment shocks can be important drivers of economic
fluctuations, there are few simple frameworks for understanding their dynamic effects. We suggest to
first estimate a second‐moment factor from the level and squared data, and then purge from it the nonlinear
variations in the level factors. Augmenting this orthogonalized second‐moment factor V1 to a
FAVAR leads to a FAVAR(s,q) model that allows the dynamic effects of first and second‐moment
shocks to be studied without direct estimation of the latent volatility processes. Our V1 is strongly
counter‐cyclical, persistent, but it not strongly correlated with the volatility of the real‐activity factor.
Its innovations explain a tangible share of the variations in housing permits, industrial production, the
fed‐funds rate, and inflation at horizons of four to five years. However, V1 does not displace other
second moment variations such as due to non‐linearity or various measures of uncertainty. The overall
finding is that second moment variations have non‐negligible macroeconomic effects and more
theorizing is needed to understand the interaction between the level and second moment dynamics.
Recent Developments in Applied Macro-econometrics
Paper Session
Saturday, Jan. 6, 2018 2:30 PM - 4:30 PM
- Chair: Barbara Rossi, ICREA-Pompeu Fabra University and Barcelona GSE
Modeling Time-varying Uncertainty of Multiple-horizon Forecast Errors
Abstract
We develop uncertainty measures for point forecasts from surveys such as theSurvey of Professional Forecasters, Blue Chip, or the Federal Open Market
Committee's Summary of Economic Projections. At a given point of time, these
surveys provide forecasts for macroeconomic variables at multiple
horizons. To track time-varying uncertainty in the associated forecast
errors, we derive a multiple-horizon specification of stochastic volatility.
Compared to constant-variance approaches, our stochastic-volatility
model improves the accuracy of uncertainty measures for survey forecasts.
Level and Volatility Factors in Macroeconomic Data
Abstract
It is well documented that the most representative panels of macroeconomic time series have a factorstructure. The conventional wisdom is to associate the common shocks with innovations to the level of
economic fundamentals. While second‐moment shocks can be important drivers of economic
fluctuations, there are few simple frameworks for understanding their dynamic effects. We suggest to
first estimate a second‐moment factor from the level and squared data, and then purge from it the nonlinear
variations in the level factors. Augmenting this orthogonalized second‐moment factor V1 to a
FAVAR leads to a FAVAR(s,q) model that allows the dynamic effects of first and second‐moment
shocks to be studied without direct estimation of the latent volatility processes. Our V1 is strongly
counter‐cyclical, persistent, but it not strongly correlated with the volatility of the real‐activity factor.
Its innovations explain a tangible share of the variations in housing permits, industrial production, the
fed‐funds rate, and inflation at horizons of four to five years. However, V1 does not displace other
second moment variations such as due to non‐linearity or various measures of uncertainty. The overall
finding is that second moment variations have non‐negligible macroeconomic effects and more
theorizing is needed to understand the interaction between the level and second moment dynamics.
Tempered and Conditionally-Optimal Particle Filtering
Abstract
Dynamic stochastic general equilibrium (DSGE) models are widely used for empirical research in macroeconomics, as well as for forecasting and quantitative policy analysis in central banks. Increasingly, these models are solved by nonlinear techniques, such as perturbation or projection methods. This talk covers recent advances in the use of particle filters to approximate the likelihood function. We discuss the tempered particle filter proposed in Herbst and Schorfheide (2017) and a conditionally-optimal particle filter (ongoing research by Aruoba, Cuba-Borda, and Schorfheide) for DSGE models with occasionally-binding constraints that are solved by constructing piecewise-linear continuous approximations to the agents' decision rules.JEL Classifications
- A1 - General Economics