« Back to Results

Developments in Macro and Forecasting

Paper Session

Saturday, Jan. 8, 2022 10:00 AM - 12:00 PM (EST)

Hosted By: Society for Nonlinear Dynamics and Econometrics
  • Chair: Tatevik Sekhposyan, Texas A&M University

A Randomized Missing Data Approach to Robust Filtering and Forecasting

Dobrislav Dobrev
,
Federal Reserve Board
Derek Hansen
,
University of Michigan
Pawel J. Szerszen
,
Federal Reserve Board

Abstract

We put forward a simple new randomized missing data (RMD) approach to robust filtering of state-space models, motivated by the idea that the inclusion of only a small fraction of available highly precise measurements can still extract most of the attainable efficiency gains for filtering latent states, estimating model parameters, and producing out-of-sample forecasts. In our general RMD framework we develop two alternative implementations: endogenous (RMD-N) and exogenous (RMD-X) randomization of missing data. A degree of robustness to outliers and model misspeci cation is achieved by purposely randomizing over the utilized subset of seemingly highly precise but possibly misspeci ed or outlier contaminated data measurements in their original time series order, while treating the rest as if missing. Time-series dependence is thus fully preserved and all available measurements can get utilized subject to a degree
of downweighting depending on the loss function of interest. The arising robustness-efficiency trade-off is controlled by varying the fraction of randomly utilized measurements or the incurred relative efficiency loss. As an empirical illustration, we show consistently attractive performance of our RMD framework in popular unobserved components models for extracting in inflation trends. We further consider model extensions that more directly reflect inflation targeting by central banks and reveal its effectiveness through improved inflation forecasting.

Anchored Inflation Expectations and the Slope of the Phillips Curve

Peter Lihn Jorgensen
,
Copenhagen Business School
Kevin J. Lansing
,
Federal Reserve Bank of San Francisco

Abstract

We estimate a New Keynesian Phillips curve that allows for changes in the degree of anchoring of agents’ subjective inflation forecasts. The estimated slope coefficient in U.S. data is highly significant and stable over the period 1960 to 2019. Out-of-sample forecasts with the model resolve both the "missing disinflation puzzle" during the Great Recession and "missing inflation puzzle" during the subsequent recovery. Using a simple New Keynesian model, we show that if agents solve a signal extraction
problem to disentangle temporary versus permanent shocks to inflation, then an increase in the policy rule coefficient on inflation serves to endogenously anchor agents inflation forecasts. Improved anchoring reduces the correlation between changes in inflation and the output gap, making the backward-looking Phillips curve appear flatter. But at the same time, improved anchoring increases the correlation between the level of inflation and the output gap, leading to a resurrection of "original" Phillips curve. Both
model predictions are consistent with U.S. data since the late 1990s.

Climate Risk and Commodity Currencies

Vegard H. Larsen
,
Norges Bank
Felix Kapfhammer
,
BI Norwegian Business School
Leif Anders Thorsrud
,
BI Norwegian Business School

Abstract

The positive relationship between real exchange rates and natural resource income is well understood and studied. However, climate change and the transition to a lower-carbon economy now challenges this relationship. We document this by proposing a novel news media-based measure of climate change transition risk and show that when such risk is high, major commodity currencies experience a persistent depreciation and the relationship between commodity price fluctuations and currencies tends to become weaker.

Nonlinear Search and Matching Explained

Nathaniel A. Throckmorton
,
College of William and Mary
Joshua Bernstein
,
Indiana University
Alexander W. Richter
,
Federal Reserve Bank of Dallas

Abstract

Prevailing wisdom suggests the matching function choice is innocuous in search and matching models. We show this is not the case. Using a closed-form global solution, we derive conditions under which the matching function generates nonlinear dynamics. Gross complementarity between vacancies and unemployed workers generates procyclical variation in the matching elasticity and negatively skewed job finding rate dynamics.
Quantitatively, this implies that the Den Haan et al. (2000) matching function more than doubles the skewness of unemployment and welfare cost of business cycles compared to the Cobb-Douglas matching function. However, the data supports the latter specification.
Discussant(s)
Leland Farmer
,
University of Virginia
Sophocles Mavroeidis
,
University of Oxford
Chiara Scotti
,
Federal Reserve Board
Nicolas Petrosky-Nadeau
,
Federal Reserve Bank of San Francisco
JEL Classifications
  • E5 - Monetary Policy, Central Banking, and the Supply of Money and Credit
  • C4 - Econometric and Statistical Methods: Special Topics