nep-ets New Economics Papers
on Econometric Time Series
Issue of 2025–02–24
six papers chosen by
Jaqueson K. Galimberti, Asian Development Bank


  1. Misspecification-Robust Shrinkage and Selection for VAR Forecasts and IRFs By Oriol Gonzalez-Casasus; Frank Schorfheide
  2. Comparing External and Internal Instruments for Vector Autoregressions By Martin Bruns; Helmut Lutkepohl
  3. Detecting Sparse Cointegration By Jesus Gonzalo; Jean-Yves Pitarakis
  4. Online Generalized Method of Moments for Time Series By Man Fung Leung; Kin Wai Chan; Xiaofeng Shao
  5. Boosting GMM with Many Instruments When Some Are Invalid and/or Irrelevant By Hao Hao; Tae-Hwy Lee
  6. Fiscal and External Sustainability: a Two-Step Time-varying Granger Causality Assessment By António Afonso; José Alves; José Carlos Coelho; Jamel Saadaoui

  1. By: Oriol Gonzalez-Casasus (University of Pennsylvania); Frank Schorfheide (University of Pennsylvania CEPR, PIER, NBER)
    Abstract: VARs are often estimated with Bayesian techniques to cope with model dimensionality. The posterior means define a class of shrinkage estimators, indexed by hyperparameters that determine the relative weight on maximum likelihood estimates and prior means. In a Bayesian setting, it is natural to choose these hyperparameters by maximizing the marginal data density. However, this is undesirable if the VAR is misspecified. In this paper, we derive asymptotically unbiased estimates of the multi-step forecasting risk and the impulse response estimation risk to determine hyperparameters in settings where the VAR is (potentially) misspecified. The proposed criteria can be used to jointly select the optimal shrinkage hyperparameter, VAR lag length, and to choose among different types of multi-step-ahead predictors; or among IRF estimates based on VARs and local projections. The selection approach is illustrated in a Monte Carlo study and an empirical application.
    Keywords: Forecasting, Hyperparameter Selection, Local Projections, Misspecification, Multi-step Estimation, Shrinkage Estimators, Vector Autoregressions
    JEL: C11 C32 C52 C53
    Date: 2025–02–05
    URL: https://d.repec.org/n?u=RePEc:pen:papers:25-003
  2. By: Martin Bruns (School of Economics, University of East Anglia); Helmut Lutkepohl (DIW Berlin and Freie Universitat Berlin)
    Abstract: In conventional proxy VAR analysis, the shocks of interest are identified by external instruments. This is typically accomplished by considering the covariance of the instruments and the reduced-form residuals. Alternatively, the instruments may be internalized by augmenting the VAR process by the instruments or proxies. These alternative identification methods are compared and it is shown that the resulting shocks obtained with the alternative approaches differ in general. Conditions are provided under which their impulse responses are nevertheless identical. If the conditions are satisfied, identification of the shocks is ensured without further assumptions. Empirical examples illustrate the results and the virtue of using the identification conditions derived in this study.
    Keywords: Structural vector autoregression, proxy VAR, augmented VAR, fundamental shocks, invertible VAR
    JEL: C32
    Date: 2025–02
    URL: https://d.repec.org/n?u=RePEc:uea:ueaeco:2025-01
  3. By: Jesus Gonzalo; Jean-Yves Pitarakis
    Abstract: We propose a two-step procedure to detect cointegration in high-dimensional settings, focusing on sparse relationships. First, we use the adaptive LASSO to identify the small subset of integrated covariates driving the equilibrium relationship with a target series, ensuring model-selection consistency. Second, we adopt an information-theoretic model choice criterion to distinguish between stationarity and nonstationarity in the resulting residuals, avoiding dependence on asymptotic distributional assumptions. Monte Carlo experiments confirm robust finite-sample performance, even under endogeneity and serial correlation.
    Date: 2025–01
    URL: https://d.repec.org/n?u=RePEc:arx:papers:2501.13839
  4. By: Man Fung Leung; Kin Wai Chan; Xiaofeng Shao
    Abstract: Online learning has gained popularity in recent years due to the urgent need to analyse large-scale streaming data, which can be collected in perpetuity and serially dependent. This motivates us to develop the online generalized method of moments (OGMM), an explicitly updated estimation and inference framework in the time series setting. The OGMM inherits many properties of offline GMM, such as its broad applicability to many problems in econometrics and statistics, natural accommodation for over-identification, and achievement of semiparametric efficiency under temporal dependence. As an online method, the key gain relative to offline GMM is the vast improvement in time complexity and memory requirement. Building on the OGMM framework, we propose improved versions of online Sargan--Hansen and structural stability tests following recent work in econometrics and statistics. Through Monte Carlo simulations, we observe encouraging finite-sample performance in online instrumental variables regression, online over-identifying restrictions test, online quantile regression, and online anomaly detection. Interesting applications of OGMM to stochastic volatility modelling and inertial sensor calibration are presented to demonstrate the effectiveness of OGMM.
    Date: 2025–02
    URL: https://d.repec.org/n?u=RePEc:arx:papers:2502.00751
  5. By: Hao Hao (Global Data Insight & Analytics, Ford Motor Company); Tae-Hwy Lee (Department of Economics, University of California Riverside)
    Abstract: When the endogenous variable is an unknown function of observable instruments, its conditional mean can be approximated using the sieve functions of observable instruments. We propose a novel instrument selection method, Double-criteria Boosting (DB), that consistently selects only valid and relevant instruments from a large set of candidate instruments. In the Monte Carlo simulation, we compare GMM using DB (DB-GMM) with other estimation methods and demonstrate that DB-GMM gives lower bias and RMSE. In the empirical application to the automobile demand, the DB-GMM estimator is suggesting a more elastic estimate of the price elasticity of demand than the standard 2SLS estimator.
    Keywords: Causal inference with high dimensional instruments; Irrelevant instruments; Invalid instruments; Instrument Selection; Machine Learning; Boosting.
    JEL: C1 C5
    Date: 2025–02
    URL: https://d.repec.org/n?u=RePEc:ucr:wpaper:202504
  6. By: António Afonso; José Alves; José Carlos Coelho; Jamel Saadaoui
    Abstract: We implement a two-step analysis of fiscal and external causality patterns using a data set covering the 27 EU countries in the period 2002Q1-2023Q4. In the 1st step, we compute fiscal and external sustainability time-varying coefficients, modelling the cointegration relationship between government revenues and government spending, and between exports and imports. In the 2nd step, we use three recursive strategies, combined with Granger causality tests: forward expanding, rolling, and recursive window methods to capture causal relationships. Our results show that: (i) peripheral countries have lower sustainability coefficients, while non-Eurozone countries have higher sustainability coefficients, (ii) after the 2008 global financial crisis, there was an improvement in fiscal and external sustainability for most countries, (iii) during the Eurozone crisis in 2010-2012, in Austria, France, Greece, Ireland, Netherlands, Slovakia and Spain, there was causality between fiscal and external sustainability, (iv) during that period, causality was observed between the external and fiscal sustainability in EMU countries (Austria, Germany, Malta, Netherlands, Slovakia, Slovenia, Spain) and in non-EMU countries.
    Keywords: fiscal sustainability; external sustainability; European Union; time-varying causality; lag-augmented vector autoregression.
    JEL: C22 C23 F32 F41 H30 H62
    Date: 2025–02
    URL: https://d.repec.org/n?u=RePEc:ise:remwps:wp03692025

This nep-ets issue is ©2025 by Jaqueson K. Galimberti. 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 https://nep.repec.org. For comments please write to the director of NEP, Marco Novarese at <director@nep.repec.org>. 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.