I try to make my work publicly available and reproducible. As publishing an open-access paper in a journal can be prohibitively expensive, please feel free to contact me for a personal copy of my papers. Please avoid paying any unfair fees to scientific publishers, see e.g. Projekt DEAL for more details.
Identification of DSGE models: a review
Abstract coming soonPaper (coming soon), Code (coming soon)
Fiscal and monetary policy in an estimated DSGE model with rare disaster
Abstract coming soonPaper (coming soon), Code (coming soon)
The effect of observables, functional specifications, model features and shocks on identification in linearized DSGE models (with Sergey Ivashchenko)Economic Modelling, in press, 2019.
The decisions a researcher makes at the model building stage are crucial for parameter identification. This paper contains a number of applied tips for solving identifiability problems and improving the strength of DSGE model parameter identification by fine-tuning the (1) choice of observables, (2) functional specifications, (3) model features and (4) choice of structural shocks. We offer a formal approach based on well-established diagnostics and indicators to uncover and address both theoretical (yes/no) identifiability issues and weak identification from a Bayesian perspective. The concepts are illustrated by two exemplary models that demonstrate the identification properties of different investment adjustment cost specifications and output-gap definitions.Online Access (DOI), Replication Code on Github, CQE Working Paper 83
Higher-order statistics for DSGE modelsEconometrics and Statistics, Volume 6, April 2018, Pages 44-56.
Closed-form expressions for unconditional moments, cumulants and polyspectra of order higher than two are derived for non-Gaussian or nonlinear (pruned) solutions to DSGE models. Apart from the existence of moments and white noise property no distributional assumptions are needed. The accuracy and utility of the formulas for computing skewness and kurtosis are demonstrated by three prominent models: the baseline medium-sized New Keynesian model used for empirical analysis (first-order approximation), a small-scale monetary business cycle model (second-order approximation) and the neoclassical growth model (third-order approximation). Both the Gaussian as well as Student's t-distribution are considered as the underlying stochastic processes. Lastly, the efficiency gain of including higher-order statistics is demonstrated by the estimation of a RBC model within a Generalized Method of Moments framework.Online Access (DOI), Online Appendix, Replication Code on Github, SFB823 Discussion Paper 4816, CQE Working Paper 43
Identification of DSGE models - The effect of higher-order approximation and pruningJournal of Economic Dynamics and Control, Volume 56, July 2015, Pages 34-54.
This paper shows how to check rank criteria for a local identification of nonlinear DSGE models, given higher-order approximations and pruning. This approach imposes additional restrictions on (higher-order) moments and polyspectra, which can be used to identify parameters that are unidentified in a first-order approximation. The identification procedures are demonstrated by means of the Kim (2003) and the An and Schorfheide (2007) models. Both models are identifiable with a second-order approximation. Furthermore, analytical derivatives of unconditional moments, cumulants and corresponding polyspectra up to fourth order are derived for the pruned state-space.Online Access (DOI), Replication Code and Additional Material on Github, CQE Working Paper 33
Permanent Working Papers
Computing parameter derivatives of policy functions solved by k-order perturbation with Dynare: With applications to identification and GMMDynare Working Paper, No. 52, 2019.
This paper presents a general procedure and recursive algorithm for analytically evaluating the derivatives of arbitrary k-order perturbation solution matrices with respect to model parameters. To this end, we unfold the k-order tensor formulas used by Dynare into an equivalent matrix representation using permutation matrices and take the derivative with respect to each model parameter separately. We show that the mathematical and computational problem is a linear, albeit large one, as it involves either solving generalized Sylvester equations or taking inverses of highly sparse matrices. The availability of parameter derivatives at any order k is beneficial in terms of easier and more reliable analysis of sensitivity and identifiability as well as more efficient estimation methods. We provide two applications: identification as well as GMM estimation at higher-order.Dynare WP52, Merge Request
A note on solving the functional equivalence between intertemporal and multisectoral investment adjustment costs (with Sergey Ivashchenko)SFB 823 Discussion Paper, No. 76/2016, 2016.
Kim (2003, JEDC 27, pp. 533-549) shows functional equivalence between intertemporal and multisectoral investment adjustments costs in a log-linearized RBC model. We provide two strategies to solve this equivalence. First, the equivalence does not hold when intertemporal adjustment costs are specified in growth rates rather than in levels. Second, the level specification can be identified with a second-order approximation of the model solution. We estimate the quadratic approximation using two extended Kalman filters within a Bayesian framework. Our estimation results confirm that both parameters are estimable in finite samples. Moreover, we provide further evidence on the stabilizing effect of pruning on the estimation algorithm.SFB823 DP7616, Estimation Results
Local identification of nonlinear and non-Gaussian DSGE models (PhD Thesis)Wissenschaftliche Schriften der WWU Münster, Reihe IV, Band 10, 2016, Paperback, ISBN: 978-3-8405-0135.
This thesis adds to the literature on the local identification of nonlinear and non-Gaussian DSGE models. It gives applied researchers a strategy to detect identification problems and means to avoid them in practice. A comprehensive review of existing methods for linearized DSGE models is provided and extended to include restrictions from higher-order moments, cumulants and polyspectra. Another approach, established in this thesis, is to consider higher-order approximations. Formal rank criteria for a local identification of the deep parameters of nonlinear or non-Gaussian DSGE models, using the pruned state-space system are derived. The procedures can be implemented prior to estimating the nonlinear model. In this way, the identifiability of the Kim (2003) and the An and Schorfheide (2007) model are demonstrated, when solved by a second-order approximation.Full Online Access