nep-ets New Economics Papers
on Econometric Time Series
Issue of 2018‒07‒09
sixteen papers chosen by
Jaqueson K. Galimberti
KOF Swiss Economic Institute

  1. Determining the dimension of factor structures in non-stationary large datasets By Matteo Barigozzi; Lorenzo Trapani
  2. Estimation of the common component in Dynamic Factor Models By Peña Sánchez de Rivera, Daniel; Caro Navarro, Ángela
  3. Factor models for portfolio selection in large dimensions: the good, the better and the ugly By Gianluca De Nard; Olivier Ledoit; Michael Wolf
  4. Model Selection in Time Series Analysis: Using Information Criteria as an Alternative to Hypothesis Testing By R. Scott Hacker; Abdulnasser Hatemi-J
  5. Stochastic model specification in Markov switching vector error correction models By Florian Huber; Michael Pfarrhofer; Thomas O. Z\"orner
  6. Testing Cointegrating Relationships Using Irregular and Non-Contemporaneous Series with an Application to Paleoclimate Data By J. Isaac Miller
  7. Semiparametrically Point-Optimal Hybrid Rank Tests for Unit Roots By Bo Zhou; Ramon van den Akker; Bas J. M. Werker
  8. Combining sign and parametric restrictions in SVARs by Givens Rotations By Lance A. Fisher; Hyeon-seung Huh
  9. An IV framework for combining sign and long?run parametric restrictions in SVARs By Lance A. Fisher; Hyeon?seung Huh
  10. Oil currencies in the face of oil shocks: what can be learned from time-varying specifications? By Jean-Pierre Allegret; Cécile Couharde; Valérie Mignon; Tovonony Razafindrabe
  11. Dynamic Effects of the Chilean Fiscal Policy By Antonio Lemus
  12. The long run impact of foreign direct investment, exports, imports and GDP: evidence for Spain from an ARDL approach By Verónica Cañal-Fernández; Julio Tascón Fernández
  13. FDI as a contributing factor to economic growth in Burkina Faso: How true is this? By Zandile, Zezethu; Phiri, Andrew
  14. An Empirical Evidence of Dynamic Interaction among price level, interest rate, money supply and real income: The case of the Indian Economy. By Rasool, Haroon; Adil, Masudul Hasan; Tarique, Md
  15. Health and income: testing for causality on European elderly people By Amélie Adeline; Eric Delattre
  16. Are long-run output growth rates falling? By Ivan Mendieta-Munoz; Mengheng Li

  1. By: Matteo Barigozzi; Lorenzo Trapani
    Abstract: We propose a procedure to determine the dimension of the common factor space in a large, possibly non-stationary, dataset. Our procedure is designed to determine whether there are (and how many) common factors (i) with linear trends, (ii) with stochastic trends, (iii) with no trends, i.e. stationary. Our analysis is based on the fact that the largest eigenvalues of a suitably scaled covariance matrix of the data (corresponding to the common factor part) diverge, as the dimension $N$ of the dataset diverges, whilst the others stay bounded. Therefore, we propose a class of randomised test statistics for the null that the $p$-th eigenvalue diverges, based directly on the estimated eigenvalue. The tests only requires minimal assumptions on the data, and no restrictions on the relative rates of divergence of $N$ and $T$ are imposed. Monte Carlo evidence shows that our procedure has very good finite sample properties, clearly dominating competing approaches when no common factors are present. We illustrate our methodology through an application to US bond yields with different maturities observed over the last 30 years. A common linear trend and two common stochastic trends are found and identified as the classical level, slope and curvature factors.
    Date: 2018–06
  2. By: Peña Sánchez de Rivera, Daniel; Caro Navarro, Ángela
    Abstract: One of the most effective techniques that allows a low-dimensional representation of Big Datasets is the Dynamic Factor Model (DFM). We analyze the finite sample performance of the well-known Principal Component estimator for the common component under different scenarios. Simulation results show that for data samples with large number of observations and small time series dimension, the variance-covariance matrix specification with lags provides better estimations than the classic variance-covariance matrix. However, in high-dimension data samples the classic variance-covariance matrix performs better no matter the sample size. Second, we apply the Principal Component estimator to obtain estimates of the business cycles of the Euro Area and its country members. This application, together with a cluster analysis, studies the phenomenon known as the Two-Speed Europe with two groups of countries not geographically related.
    Keywords: Time series ; Factor Models ; Principal Components ; Canonical Correlations
    Date: 2018–06–15
  3. By: Gianluca De Nard; Olivier Ledoit; Michael Wolf
    Abstract: This paper injects factor structure into the estimation of time-varying, large-dimensional covariance matrices of stock returns. Existing factor models struggle to model the covariance matrix of residuals in the presence of conditional heteroskedasticity in large universes. Conversely, rotation-equivariant estimators of large-dimensional time-varying covariance matrices forsake directional information embedded in market-wide risk factors. We introduce a new covariance matrix estimator that blends factor structure with conditional heteroskedasticity of residuals in large dimensions up to 1000 stocks. It displays superior all-around performance on historical data against a variety of state-of-the-art competitors, including static factor models, exogenous factor models, sparsity-based models, and structure-free dynamic models. This new estimator can be used to deliver more efficient portfolio selection and detection of anomalies in the cross-section of stock returns.
    Keywords: Dynamic conditional correlations, factor models, multivariate GARCH, Markowitz portfolio selection, nonlinear shrinkage
    JEL: C13 C58 G11
    Date: 2018–06
  4. By: R. Scott Hacker; Abdulnasser Hatemi-J
    Abstract: The issue of model selection in applied research is of vital importance. Since the true model in such research is not known, which model should be used from among various potential ones is an empirical question. There might exist several competitive models. A typical approach to dealing with this is classic hypothesis testing using an arbitrarily chosen significance level based on the underlying assumption that a true null hypothesis exists. In this paper we investigate how successful this approach is in determining the correct model for different data generating processes using time series data. An alternative approach based on more formal model selection techniques using an information criterion or cross-validation is suggested and evaluated in the time series environment via Monte Carlo experiments. This paper also explores the effectiveness of deciding what type of general relation exists between two variables (e.g. relation in levels or relation in first differences) using various strategies based on hypothesis testing and on information criteria with the presence or absence of unit roots.
    Date: 2018–05
  5. By: Florian Huber; Michael Pfarrhofer; Thomas O. Z\"orner
    Abstract: This paper proposes a hierarchical modeling approach to perform stochastic model specification in Markov switching vector error correction models. We assume that a common distribution gives rise to the regime-specific regression coefficients. The mean as well as the variances of this distribution are treated as fully stochastic and suitable shrinkage priors are used. These shrinkage priors enable to assess which coefficients differ across regimes in a flexible manner. In the case of similar coefficients, our model pushes the respective regions of the parameter space towards the common distribution. This allows for selecting a parsimonious model while still maintaining sufficient flexibility to control for sudden shifts in the parameters, if necessary. In the empirical application, we apply our modeling approach to Euro area data and assume that transition probabilities between expansion and recession regimes are driven by the cointegration errors. Our findings suggest that lagged cointegration errors have predictive power for regime shifts and these movements between business cycle stages are mostly driven by differences in error variances.
    Date: 2018–07
  6. By: J. Isaac Miller (Department of Economics, University of Missouri, Columbia, Missouri, USA)
    Abstract: Time series that are observed neither regularly nor contemporaneously pose problems for most multivariate analyses. Common and intuitive solutions to these problems include linear and step interpolation or other types of imputation to a higher, regular frequency. However, interpolation is known to cause serious problems with the size and power of statistical tests. Due to the difficulty in measuring stochastically varying paleoclimate phenomena such as CO2 concentrations and surface temperatures, time series of such measurements are observed neither regularly nor contemporaneously. This paper presents large- and small-sample analyses of the size and power of cointegration tests of time series with these features and confirms the robustness of cointegration of these two series found in the extant literature. Step interpolation is preferred over linear interpolation.
    Keywords: cointegration, irregularly time series, non-contemporaneous time series, misaligned time series, paleoclimate data
    JEL: C12 C22 C32 Q54
    Date: 2018–06–29
  7. By: Bo Zhou; Ramon van den Akker; Bas J. M. Werker
    Abstract: We propose a new class of unit root tests that exploits invariance properties in the Locally Asymptotically Brownian Functional limit experiment associated to the unit root model. The invariance structures naturally suggest tests that are based on the ranks of the increments of the observations, their average, and an assumed reference density for the innovations. The tests are semiparametric in the sense that they are valid, i.e., have the correct (asymptotic) size, irrespective of the true innovation density. For a correctly specified reference density, our test is point-optimal and nearly efficient. For arbitrary reference densities, we establish a Chernoff-Savage type result, i.e., our test performs as well as commonly used tests under Gaussian innovations but has improved power under other, e.g., fat-tailed or skewed, innovation distributions. To avoid nonparametric estimation, we propose a simplified version of our test that exhibits the same asymptotic properties, except for the Chernoff-Savage result that we are only able to demonstrate by means of simulations.
    Date: 2018–06
  8. By: Lance A. Fisher (Macquarie University); Hyeon-seung Huh (Yonsei University)
    Abstract: This paper develops a method for combining sign and parametric restrictions in SVARs by means of Givens matrices. The Givens matrix is used to rotate an initial set of orthogonal shocks in the SVAR. Parametric restrictions are imposed on the Givens matrix in a manner which utilises its properties. This gives rise to a system of equations which can be solved recursively for the ¡®angles¡¯ in the constituent Givens matrices to enforce the parametric restrictions. The method is applied to several identifications which involve a combination of sign restrictions, and long-run and/or contemporaneous restrictions in Peersman¡¯s (2005) SVAR for the US economy. The method is compared to the recently developed method of Aries, Rubio-Ramirez and Waggoner (2018) which combines zero and sign restrictions.
    Keywords: structural vector autoregressions, sign and parametric restrictions, Givens rotations, QR decomposition
    JEL: C32 C51 E32
    Date: 2018–06
  9. By: Lance A. Fisher (Macquarie University); Hyeon?seung Huh (Yonsei University)
    Abstract: This paper develops a method to impose a long?run restriction in an instrumental variables (IV) framework in a SVAR which is comprised of both I(1) and I(0) variables when the shock associated with one of the I(0) variables is made transitory. This is the identification which is utilized in the small open economy SVAR that we take from the literature. The method is combined with a recently developed sign restrictions approach which can be applied in an IV setting. We then consider an alternate identification in this SVAR which makes the shocks associated with all of the I(0) variables transitory. In this case, we show that another method can be used to impose the long?run restrictions. The results from both methods are reported for the SVARs estimated with Canadian data.
    Keywords: sign restrictions, long?run parametric restrictions, IV estimation, algorithms, generated coefficients, small open economy, Canada
    JEL: C32 C36 C51 F41
    Date: 2018–07
  10. By: Jean-Pierre Allegret (EconomiX - UPN - Université Paris Nanterre - CNRS - Centre National de la Recherche Scientifique); Cécile Couharde (EconomiX - UPN - Université Paris Nanterre - CNRS - Centre National de la Recherche Scientifique); Valérie Mignon (EconomiX - UPN - Université Paris Nanterre - CNRS - Centre National de la Recherche Scientifique); Tovonony Razafindrabe (EconomiX - UPN - Université Paris Nanterre - CNRS - Centre National de la Recherche Scientifique)
    Abstract: While the oil currency property is clearly establis hed from a theoretical viewpoint, its existence is less clear-cut in the empirical literature. We investigate the reasons for this apparent puzzle by studying the time-varying nature of the relationship between real effective exchange rates of five oil exporters and the real price of oil in the aftermath of the oil price shocks of the last two decades. Accordingly, we rely on a time-varying parameter VAR specification which allows the responses of real exchange rates to different oil price shocks to evolve over time. We find that the reason of the mixed results obtained in the empirical literature is that oil currencies follow different hybrid models in the sense that oil countries’real exchange rates may be driven by one or several sources of oil price shocks that furthermore can vary over time. In addition to structural changes affecting oil countries, structural changes arising from the oil market itself through the various, time-varying sources of oil price shocks are found to be crucial.
    Keywords: oil currencies,oil shocks,Time-Varying Parameter VAR model,exchange rates
    Date: 2017
  11. By: Antonio Lemus
    Abstract: In Chile, the empirical literature studying the dynamic effects of fiscal policy and fiscal multipliers, using linear vector autoregression models, disagrees on the effects of government spending and taxes on output. In this paper, we bring new elements to this debate. We include the nonlinear dimension of vector autoregression models to answer if the state, “tight” or “normal”, of the Chilean economy, affects fiscal policy effectiveness. Last, based on the nonlinear framework we question if monetary policy has an influence on the size of fiscal multipliers. We find that: (i) once using the same quarterly data, the size of fiscal multipliers not only varies depending on the identification strategy and the linear vector autoregression model used but also on the definitions of government spending and taxes considered; (ii) the government spending multiplier from the nonlinear framework differs, being about the unit in the “tight” regime and around -0.5 in the “normal” regime; (iii) government spending and tax multipliers in the nonlinear framework are smaller when monetary policy is taken into account, which influences the effectiveness of fiscal policy.
    Keywords: Fiscal Policy, Fiscal Multipliers, Vector Autoregression Models
    JEL: C11 C32 E62
    Date: 2018
  12. By: Verónica Cañal-Fernández (Department of Economics Faculty of Economics and Business University of Oviedo); Julio Tascón Fernández (Department of Economics Faculty of Economics and Business University of Oviedo)
    Abstract: This paper analyses the relationship between foreign direct investment (FDI), exports and economic growth in Spain using annual time series data for the period 1970 to 2016. To examine these linkages the autoregressive distributed lag (ARDL) bounds testing approach to cointegration for the long-run is applied. The error correction model (ECM) is used to examine the short-run dynamics and the vector error correction model (VECM) Granger causality approach is used to investigate the direction of causality. The results confirm a long-run relationship among the examined variables. The Granger causality test indicates a strong unidirectional causality between FDI and exports with direction from FDI to exports. Besides, the results for the relationship between FDI and economic growth are interesting and indicate that there is no significant Granger causality from FDI to economic growth and vice-versa.
    Keywords: Foreign direct investment; exports; imports; GDP; ARDL bounds; causality
    JEL: C22 E31 E50
    Date: 2018–04
  13. By: Zandile, Zezethu; Phiri, Andrew
    Abstract: Much emphasis has been placed on attracting FDI into Burkina Faso as a catalyst for improved economic growth within the economy. Against the lack of empirical evidence evaluating this claim, we use data collected from 1970 to 2017 to investigate the FDI-growth nexus for the country using the ARDL bounds cointegration analysis. Our empirical model is derived from endogenous growth theoretical framework in which FDI may have direct or spillover effects on economic growth via improved human capital development as well technological developments reflected in urbanization and improved export growth. Our findings fail to establish any direct or indirect effects of FDI on economic growth except for FDI’s positive interaction with export-oriented growth, albeit being constrained to the short-run. Therefore, in summing up our recommendations, political reforms and the building of stronger economic ties with the international community in order to raise investor confidence, which has been historically problematic, should be at the top of the agenda for policymakers in Burkina Faso.
    Keywords: Foreign direct investment; economic growth; Burkina Faso; West Africa; ARDL cointegration.
    JEL: C13 C32 C51 F21 O40
    Date: 2018–06–11
  14. By: Rasool, Haroon; Adil, Masudul Hasan; Tarique, Md
    Abstract: Monetary policy approaches in India has changed from simple monetary targeting framework in the mid-1980s to multiple indicator approach in the late 1990s and to the current flexible inflation targeting framework. The aim of this study is to investigate the relationship among selected macroeconomic variables such as, money supply, real income, price level and interest rate for period 1998 to 2014 in case of India; a period when the Multiple Indicator Approach (MIA) was implemented. The study employs vector autoregression (VAR) approach to examine the dynamics of the relationship between variables. The result shows that lags of all dependent variables are significant except real income. The Granger causality via VAR framework suggests that four pairs of Granger causality exist, in particular, bi-directional causality exists between money supply and price level. Interest rate Granger causes both income and price level, and lastly money supply causes the rate of interest. However, the study could not find any causal relationship between real income and money supply in either direction. The findings of Impulse response functions and Variance decomposition reinforce causality results. Finally, the estimated result supports the arguments which are made in favour of policy move from MIA to inflation targeting framework.
    Keywords: Multiple Indicator Approach, VAR, Granger causality, IRF and VD.
    JEL: C5 E4 E5
    Date: 2018
  15. By: Amélie Adeline; Eric Delattre (Université de Cergy-Pontoise, THEMA)
    Abstract: Socioeconomic status and health are positively related, also known as the "healthincome gradient". However, when considering the causal impact of income on health, the reverse causality might be at play. Income inequalities are an important factor in health inequality such that policy makers who aim at improving general health or narrowing inequalities using public policies, need to understand the sources and the direction of the causality between income and health. We thus investigate bivariate causal effects between the two by highlighting the Granger causality. Using the Survey of Health, Aging and Retirement in Europe (SHARE), we find evidence of persistent causal effects running from income to health and from health to income. Results, using a Full Information Maximum Likelihood estimator (FIML), suggest that considering a simultaneous equations approach is required because there are unobservable factors common to both equations in the individual e ects (statistically significant correlation between the two equations).
    Keywords: Granger causality; income; simultaneity; self-assessed health; FIML.
    JEL: C32 C33 D31 I10 J14
    Date: 2018
  16. By: Ivan Mendieta-Munoz; Mengheng Li
    Abstract: This paper studies the evolution of long-run output and labour productivity growth rates in the G-7 countries during the post-war period. We estimate the growth rates consistent with a constant unemployment rate using time-varying parameter models that incorporate both stochastic volatility and a Heckman-type two-step estimation procedure that deals with the possible endogeneity problem in the econometric models. The results show a significant decline in long-run growth rates that is not associated with the detrimental effects of the Great Recession, and that the rate of growth of labour productivity appears to be behind the slowdown in long-run GDP growth.
    Keywords: Long-run output growth rates, unobserved components, Kalman filter, time- varying parameter models, stochastic volatility, Heckman two-step bias correction. JEL Classification: O41, O47, C15, C32
    Date: 2018

This nep-ets issue is ©2018 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 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.