Search Results
Working Paper
Dynamic Identification Using System Projections on Instrumental Variables
We propose System Projections on Instrumental Variables (SP-IV) to estimate structural relationships using regressions of structural impulse responses obtained from local projections or vector autoregressions. Relative to IV with distributed lags of shocks as instruments, SP-IV imposes weaker exogeneity requirements and can improve efficiency and increase effective instrument strength relative to the typical 2SLS estimator. We describe inference under strong and weak identification. The SP-IV estimator outperforms other estimators of Phillips Curve parameters in simulations. We estimate the ...
Working Paper
Dynamic Identification Using System Projections on Instrumental Variables
We propose System Projections on Instrumental Variables (SP-IV) to identify structural relationships using regressions of impulse responses from local projections or vector autoregressions. Relative to 2SLS with distributed lags as instruments, SP-IV weakens exogeneity requirements and can improve efficiency and effective instrument strength. We describe inference under strong and weak identification. The SP-IV estimator outperforms other estimators of Phillips Curve parameters in simulations. We estimate the Phillips Curve implied by the main business cycle shock of Angeletos et al. (2020) ...
Working Paper
Measurement Errors and Monetary Policy: Then and Now
Should policymakers and applied macroeconomists worry about the difference between real-time and final data? We tackle this question by using a VAR with time-varying parameters and stochastic volatility to show that the distinctionbetween real-time data and final data matters for the impact of monetary policy shocks: The impact on final data is substantially and systematically different (in particular, larger in magnitude for different measures of real activity) from theimpact on real-time data. These differences have persisted over the last 40 years and should be taken into account when ...
Working Paper
The Role of the Prior in Estimating VAR Models with Sign Restrictions
Several recent studies have expressed concern that the Haar prior typically imposed in estimating sign-identified VAR models may be unintentionally informative about the implied prior for the structural impulse responses. This question is indeed important, but we show that the tools that have been used in the literature to illustrate this potential problem are invalid. Specifically, we show that it does not make sense from a Bayesian point of view to characterize the impulse response prior based on the distribution of the impulse responses conditional on the maximum likelihood estimator of ...
Working Paper
Affine term structure pricing with bond supply as factors
This paper presents a theoretical model for analyzing the effect of the maturity structure of government debt on the yield curve. It is an ATSM (affine term structure model) in which the factors for the yield curve include, in addition to the short rate, the government bond supply for each maturity. The supply shock is not restricted to be perfectly correlated across maturities. The effect on the yield curve of a bond supply shock that is local to a maturity is largest at the maturity. This hump-shaped response of the yield curve persists in spite of the absence of preferred-habitat investors.
Working Paper
Averaging Impulse Responses Using Prediction Pools
Macroeconomists construct impulse responses using many competing time series models and different statistical paradigms (Bayesian or frequentist). We adapt optimal linear prediction pools to efficiently combine impulse response estimators for the effects of the same economic shock from this vast class of possible models. We thus alleviate the need to choose one specific model, obtaining weights that are typically positive for more than one model. Three Monte Carlo simulations and two monetary shock empirical applications illustrate how the weights leverage the strengths of each model by (i) ...
Report
Micro Responses to Macro Shocks
We study panel data regression models when the shocks of interest are aggregate and possibly small relative to idiosyncratic noise. This speaks to a large empirical literature that targets impulse responses via panel local projections. We show how to interpret the estimated coefficients when units have heterogeneous responses and how to obtain valid standard errors and confidence intervals. A simple recipe leads to robust inference: including lags as controls and then clustering at the time level. This strategy is valid under general error dynamics and uniformly over the degree of ...
Working Paper
Dynamic Identification Using System Projections and Instrumental Variables
We propose System Projections on Instrumental Variables (SP-IV) to estimate dynamic structural relationships using impulse responses obtained from local projections or vector autoregressions. SP-IV replaces lag sequences of instruments in traditional IV with lead sequences of endogenous variables. By allowing the inclusion of lagged variables as controls, SP-IV weakens exogeneity requirements and can improve efficiency and effective instrument strength relative to 2SLS. We provide inference procedures under strong and weak identification, and show that SP-IV outperforms conventional IV ...
Working Paper
Uniform Priors for Impulse Responses
There has been a call for caution when using the conventional method for Bayesian inference in set-identified structural vector autoregressions on the grounds that the uniform prior over the set of orthogonal matrices could be nonuniform for key objects of interest. This paper challenges this call. Although the prior distributions of individual impulse responses induced by the conventional method may be nonuniform, they typically do not drive the posteriors if one does not condition on the reduced-form parameters. Importantly, when the focus is on joint inference, the uniform prior over the ...
Working Paper
A Sufficient Statistics Approach for Macro Policy Evaluation
The evaluation of macroeconomic policy decisions has traditionally relied on the formulation of a specific economic model. In this work, we show that two statistics are sufficient to detect, often even correct, non-optimal policies, i.e., policies that do not minimize the loss function. The two sufficient statistics are (i) the effects of policy shocks on the policy objectives, and (ii) forecasts for the policy objectives conditional on the policy decision. Both statistics can be estimated without relying on a specific model. We illustrate the method by studying US monetary policy decisions.