Joint work with Silvia Goncalves, Ana Maria Herrera, Lutz Kilian and Elena Peavento
R&R at the Journal of Business and Economic Statistics
We propose a semiparametric local projection estimator of nonlinear impulse response functions for a broad class of structural dynamic models relevant for applied macroeconomics, including models with nonlinearly transformed regressors, state dependent coefficients, and nonlinear interactions between shocks and state variables. The estimator is based on a doubly robust moment condition that identifies the average response function as a linear functional of a nonparametric conditional mean, augmented by a density ratio that captures the effect of shifting the shock of interest. We combine this moment condition with cross-fitting that handles serial dependence. The resulting estimator is root-T consistent and asymptotically normal. We examine the finite-sample performance of the estimator across a range of nonlinear data generating processes and illustrate its use in two empirical examples.
Joint work with Sílvia Gonçalves and Ana María Herrera
Disclaimer: This draft was written by me to summarize our ideas and current result, it should not be taken as a draft of the final paper. We are currently working on relaxing the i.i.d assumption on xt.
We study the distribution of the multi-horizon effects of an identified shock on endogenous variables. We introduce a novel policy parameter, the Impulse Response quantiles (IRQ), defined as the quantiles of the distribution of the difference between the observed process and the counterfactual process. We first identify the IRQ in a general non-parametric model and two widely used nonlinear parametric models: a linear structural dynamic model that includes nonlinearly transformed regressors and a state-dependent model. We propose estimation and inference methods for all three. We apply the method to the oil supply shocks of Kanzig (2021) and the monetary policy shocks of Fanelli and Marsi (2022).
This work was presented at the 2025 Canadian Economics Association Conference
Joint work with Santiago Camara
Draft available soon.
This paper presents evidence of significant global spillovers of the ECB's anti fragmentation policy. We focus on the ECB's management of fragmentation risk in the sovereign bond market, a policy through which the ECB sought to reduce the spread between interest rates in core versus periphery countries in the union. Recovering shocks using the identification strategy developed by Fanelli and Marsi (2022) and a panel of 22 Emerging and 12 Advanced Economies we show that conventional and spread policy shocks have intuitive and very different international spillovers. In particular, an increase in ECB rates due to a ``spread shock'' leads to a significant reduction in long term rates, and a persistent expansion in equity and industrial indexes.
Selected scaled impulse response functions for shocks of different sizes.
We investigate the presence of sign and size non-linearities in the impact of the European Central Bank’s Anti-Fragmentation Policy on non-ERM II, EU countries. After identifying three orthogonal monetary policy shock using the method of Fanelli and Marsi [2022], we then select an optimal specification and estimate both linear and non linear impulse response functions using local projections (Dufour and Renault [1998], Gonçalves et al. [2021]). The choice of non-linear transformations to separate sign and size effects is based on Caravello and Martinez-Bruera [Working Paper, 2024]. Lastly we compare the linear model to the non-linear ones using a battery of Wald tests and find significant evidence of sign non-linearities in the international spillovers of ECB policy.
This work was presented at the 2024 Canadian Economics Association Conference