nep-ecm New Economics Papers
on Econometrics
Issue of 2018‒05‒14
nine papers chosen by
Sune Karlsson
Örebro universitet

  1. New HSIC-based tests for independence between two stationary multivariate time series By Guochang Wang; Wai Keung Li; Ke Zhu
  2. Forecasting with High-Dimensional Panel VARs By Gary Koop; Dimitris Korobilis
  3. Adaptive Hierarchical Priors for High-Dimensional Vector Autoregressions By Dimitris Korobilis; Davide Pettenuzzo
  4. Sufficient Statistics for Unobserved Heterogeneity in Structural Dynamic Logit Models By Victor Aguirregabiria; Jiaying Gu; Yao Luo
  5. Identifying Effects of Multivalued Treatments By Sokbae Lee; Bernard Salani\'e
  6. Strong consistency of the least squares estimator in regression models with adaptive learning By Norbert Christopeit; Michael Massmann
  7. Identification of heterogeneous treatment effects as a function of potential untreated outcome under the nonignorable assignment condition By Takahiro Hoshino; Keisuke Takahata
  8. Reducing Bias in a Matching Estimation of Endogenous Treatment Effect By A. Di Pino; M.G. Campolo; E. Otranto
  9. Can News and Noise Shocks Be Disentangled? By Luca Benati

  1. By: Guochang Wang; Wai Keung Li; Ke Zhu
    Abstract: This paper proposes some novel one-sided omnibus tests for independence between two multivariate stationary time series. These new tests apply the Hilbert-Schmidt independence criterion (HSIC) to test the independence between the innovations of both time series. Under regular conditions, the limiting null distributions of our HSIC-based tests are established. Next, our HSIC-based tests are shown to be consistent. Moreover, a residual bootstrap method is used to obtain the critical values for our HSIC-based tests, and its validity is justified. Compared with the existing cross-correlation-based tests for linear dependence, our tests examine the general (including both linear and non-linear) dependence to give investigators more complete information on the causal relationship between two multivariate time series. The merits of our tests are illustrated by some simulation results and a real example.
    Date: 2018–04
  2. By: Gary Koop (Department of Economics, University of Strathclyde, UK; Rimini Centre for Economic Analysis); Dimitris Korobilis (Essex Business School, University of Essex, UK; Rimini Centre for Economic Analysis)
    Abstract: This paper develops methods for estimating and forecasting in Bayesian panel vector autoregressions of large dimensions with time-varying parameters and stochastic volatility. We exploit a hierarchical prior that takes into account possible pooling restrictions involving both VAR coefficients and the error covariance matrix, and propose a Bayesian dynamic learning procedure that controls for various sources of model uncertainty. We tackle computational concerns by means of a simulation-free algorithm that relies on an analytical approximation of the posterior distribution. We use our methods to forecast inflation rates in the eurozone and show that forecasts from our flexible specification are superior to alternative methods for large vector autoregressions.
    Keywords: Panel VAR, inflation forecasting, Bayesian, time-varying parameter model
    Date: 2018–05
  3. By: Dimitris Korobilis (Essex Business School, University of Essex, UK; Rimini Centre for Economic Analysis); Davide Pettenuzzo (Sachar International Center, Brandeis University, USA)
    Abstract: This paper proposes a simulation-free estimation algorithm for vector autoregressions (VARs) that allows fast approximate calculation of marginal parameter posterior distributions. We apply the algorithm to derive analytical expressions for independent VAR priors that admit a hierarchical representation and which would typically require computationally intensive posterior simulation methods. The benefits of the new algorithm are explored using three quantitative exercises. First, a Monte Carlo experiment illustrates the accuracy and computational gains of the proposed estimation algorithm and priors. Second, a forecasting exercise involving VARs estimated on macroeconomic data demonstrates the ability of hierarchical shrinkage priors to find useful parsimonious representations. We also show how our approach can be used for structural analysis and that it can successfully replicate important features of news-driven business cycles predicted by a large-scale theoretical model.
    Keywords: Bayesian VARs, Mixture prior, Large datasets, Macroeconomic forecasting
    JEL: C11 C13 C32 C53
    Date: 2018–05
  4. By: Victor Aguirregabiria; Jiaying Gu; Yao Luo
    Abstract: We study the identification and estimation of structural parameters in dynamic panel data logit models where decisions are forward-looking and the joint distribution of unobserved heterogeneity and observable state variables is nonparametric, i.e., fixed-effects model. We consider models with two endogenous state variables: the lagged decision variable, and the time duration in the last choice. This class of models includes as particular cases important economic applications such as models of market entry-exit, occupational choice, machine replacement, inventory and investment decisions, or dynamic demand of differentiated products. The identification of structural parameters requires a sufficient statistic that controls for unobserved heterogeneity not only in current utility but also in the continuation value of the forward-looking decision problem. We obtain the minimal sufficient statistic and prove identification of some structural parameters using a conditional likelihood approach. We apply this estimator to a machine replacement model.
    Keywords: Panel data discrete choice models; Dynamic structural models; Fixed effects; Unobserved heterogeneity; Structural state dependence; Identification; Sufficient statistic.
    JEL: C23 C25 C41 C51 C61
    Date: 2018–05–10
  5. By: Sokbae Lee; Bernard Salani\'e
    Abstract: Multivalued treatment models have typically been studied under restrictive assumptions: ordered choice, and more recently unordered monotonicity. We show how treatment effects can be identified in a more general class of models that allows for multidimensional unobserved heterogeneity. Our results rely on two main assumptions: treatment assignment must be a measurable function of threshold-crossing rules, and enough continuous instruments must be available. We illustrate our approach for several classes of models.
    Date: 2018–04
  6. By: Norbert Christopeit (University of Bonn); Michael Massmann (VU Amsterdam)
    Abstract: This paper looks at the strong consistency of the ordinary least squares (OLS) estimator in a stereotypical macroeconomic model with adaptive learning. It is a companion to Christopeit & Massmann (2017, Econometric Theory) which considers the estimator’s convergence in distribution and its weak consistency in the same setting. Under constant gain learning, the model is closely related to stationary, (alternating) unit root or explosive autoregressive processes. Under decreasing gain learning, the regressors in the model are asymptotically collinear. The paper examines, first, the issue of strong convergence of the learning recursion: It is argued that, under constant gain learning, the recursion does not converge in any probabilistic sense, while for decreasing gain learning rates are derived at which the recursion converges almost surely to the rational expectations equilibrium. Secondly, the paper establishes the strong consistency of the OLS estimators, under both constant and decreasing gain learning, as well as rates at which the estimators converge almost surely. In the constant gain model, separate estimators for the intercept and slope parameters are juxtaposed to the joint estimator, drawing on the recent literature on explosive autoregressive models. Thirdly, it is emphasised that strong consistency is obtained in all models although the near-optimal condition for the strong consistency of OLS in linear regression models with stochastic regressors, established by Lai & Wei (1982), is not always met.
    Keywords: adaptive learning, non-stationary regression, ordinary least squares, almost sure convergence
    JEL: C22 C51 D83
    Date: 2018–05–11
  7. By: Takahiro Hoshino (Department of Economics, Keio University); Keisuke Takahata (Graduate of School of Economics, Keio University)
    Abstract: We provide sufficient conditions for the identification of hetero- geneous treatment effects (HTE), in which the missing mechanism is nonignorable, when the information on the marginal distribution of untreated outcome is available. It is also shown that, under such a situ- ation, the same result holds for the identification of average treatment effects (ATE). Exposing certain additivity on the regression function of the assignment probability, we reduce the identication of HTE to the uniqueness of a solution of some integral equation, and discuss it borrowing the idea from the literature on statistical inverse prob- lems. Our result contributes to theoretical understandings in causal inference with heterogeneity and also the relaxation of the conditional independence assumption in statistical data fusion or statistical data combination.
    Keywords: nonignorable missing, causal inference, identifiability
    JEL: C13 C31 C83
    Date: 2018–04–09
  8. By: A. Di Pino; M.G. Campolo; E. Otranto
    Abstract: The traditional matching methods for the estimation of treatment parameters are often affected by selectivity bias due to the endogenous joint influence of latent factors on the assignment to treatment and on the outcome, especially in a cross-sectional framework. In this study, we show that the influence of unobserved factors involves a cross-correlation between the endogenous components of propensity scores and causal effects. A correction for the effects of this correlation on matching results leads to a reduction of bias. A Monte Carlo experiment and an empirical application using the LaLonde's experimental data set support this finding.
    Keywords: endogenous component of propensity scores;endogenous treatment;propensity score matching;State-Space Model
    Date: 2018
  9. By: Luca Benati
    Abstract: Chahrour and Jurado (2018) have shown that news and noise shocks are observationally equivalent when the econometrician only observes a fundamental process and agents’ expectations about it. We show that the observational equivalence result no longer holds when the econometrician observes a fundamental process and a noisy signal of it. Working with an RBC model with noise about TFP, we further show that, even if the signal is not directly observed by the econometrician, it can be inferred through its impact on other macroeconomic variables, since they are optimally chosen by agents conditional on all information, including the signal itself. In particular, we show that under these circumstances news and noise shocks can be exactly recovered in population. Our results demonstrate that news and noise shocks are not observationally equivalent for an econometrician exploiting all the information contained in standard macroeconomic time series.
    Date: 2018–05

This nep-ecm issue is ©2018 by Sune Karlsson. It is provided as is without any express or implied warranty. It may be freely redistributed in whole or in part for any purpose. If distributed in part, please include this notice.
General information on the NEP project can be found at For comments please write to the director of NEP, Marco Novarese at <>. Put “NEP” in the subject, otherwise your mail may be rejected.
NEP’s infrastructure is sponsored by the School of Economics and Finance of Massey University in New Zealand.