nep-ecm New Economics Papers
on Econometrics
Issue of 2006‒01‒01
eleven papers chosen by
Sune Karlsson
Orebro University

  1. Panel Data Unit Roots Tests: The Role of Serial Correlation and the Time Dimension By Stefan De Wachter; Richard D.F. Harris; Elias Tzavalis
  2. A Testing Procedure for Determining the Number of Factors in Approximate Factor Models with Large Datasets By George Kapetanios
  3. Tests Against Stationary and Explosive Alternatives in Vector Autoregressive Models By Ahlgren, Niklas; Nyblom, Jukka
  4. Panel Cointegration Testing in the Presence of Common Factors By Gengenbach,Christian; Palm,Franz C.; Urbain,Jean-Pierre
  5. Parametric Binary Choice Models By Michael Lechner; Stefan Lollivier; Thierry Magnac
  6. Purchasing Power Parity and Heterogeneous Mean Reversion By Koedijk, C.G.; Tims, B.; Dijk, M.A. van
  7. Missing data in optimal scaling By Pieralda FERRARI; Paola ANNONI
  8. The Bias of Elasticity Estimtors in Linear Regression: Some Analytic Results By Chen Qian; David E. Giles
  9. The Analysis of Location, Co-location and Urbanisation Economics By Johansson, Börje; Forslund, Ulla
  10. Markov-switching structural vector autoregressions: theory and application By Juan Francisco Rubio-Ramírez; Daniel Waggoner; Tao Zha
  11. Clustering of auto supplier plants in the U.S.: GMM spatial logit for large samples By Thomas H. Klier; Dan McMillen

  1. By: Stefan De Wachter (University of Oxford); Richard D.F. Harris (University of Exeter); Elias Tzavalis (Queen Mary, University of London)
    Abstract: We investigate the influence of residual serial correlation and of the time dimension on statistical inference for a unit root in dynamic longitudinal data, known as panel data in econometrics. To this end, we introduce two test statistics based on method of moments estimators. The first is based on the generalised method of moments estimators, while the second is based on the instrumental variables estimator. Analytical results for the IV based test in a simplified setting show that (i) large time dimension panel unit root tests will suffer from serious size distortions in finite samples, even for samples that would normally be considered large in practice, and (ii) negative serial correlation in the error terms of the panel reduces the power of the unit root tests, possibly up to a point where the test becomes biased. However, near the unit root the test is shown to have power against a wide range of alternatives. These findings are confirmed in a more general set-up through a series of Monte Carlo experiments.
    Keywords: Dynamic longitudinal (panel) data, Generalized method of moments, Instrumental variables, Unit roots, Moving average errors
    JEL: C22 C23
    Date: 2005–12
  2. By: George Kapetanios (Queen Mary, University of London)
    Abstract: The paradigm of a factor model is very appealing and has been used extensively in economic analyses. Underlying the factor model is the idea that a large number of economic variables can be adequately modelled by a small number of indicator variables. Throughout this extensive research activity on large dimensional factor models a major preoccupation has been the development of tools for determining the number of factors needed for modelling. This paper provides builds on the work of Kapetanios (2004) to provide an alternative method to information criteria as a tool for estimating the number of factors in large dimensional factor models. The new method is robust to considerable cross-sectional and temporal dependence. The theoretical properties of the method are explored and an extensive Monte Carlo study is undertaken. Results are favourable for the new method and suggest that it is a reasonable alternative to existing methods.
    Keywords: Factor models, Large sample covariance matrix, Maximum eigenvalue
    JEL: C12 C15 C23
    Date: 2005–12
  3. By: Ahlgren, Niklas (Swedish School of Economics and Business Administration); Nyblom, Jukka (University of Joensuu)
    Abstract: The article proposes new tests for the number of unit roots in vector autoregressive models based on the eigenvalues of the companion matrix. Both stationary and explosive alternatives are considered. The limiting distributions of test statistics depend only on the number of unit roots. Size and power are investigated and it is found that the new test against stationary alternatives compares favorably with the widely used likelihood ratio test for the cointegrating rank. The powers are prominently higher against explosive than stationary alternatives. Some empirical examples are provided to show how to use the new tests with real data.
    Keywords: Asymptotic local power; Cointegration; Companion matrix; Unit root
    Date: 2005–12–14
  4. By: Gengenbach,Christian; Palm,Franz C.; Urbain,Jean-Pierre (METEOR)
    Abstract: Panel unit root and no-cointegration tests that rely on cross-sectional independence of the panel unit experience severe size distortions when this assumption is violated, as has e.g. been shown by Banerjee, Marcellino and Osbat (2004, 2005) via Monte Carlo simulations. Several studies have recently addressed this issue for panel unit root test using a common factor structure to model the cross-sectional dependence, but not much work has been done yet for panel no-cointegration tests. This paper proposes a model for panel no-cointegration using an unobserved common factor structure, following the work on Bai and Ng (2004) for panel unit roots. The model enables us to distinguish two important cases: (i) the case when the non-stationarity in the data is driven by a reduced number of common stochastic trends, and (ii) the case where we have common and idiosyncratic stochastic trends present in the data. We study the asymptotic behavior of some existing, residual-based panel no-cointegration, as suggested by Kao (1999) and Pedroni (1999, 2004). Under the DGP used, the test statistics are no longer asymptotically normal, and convergence occurs at rate T rather than sqrt(N)T as for independent panels. We then examine the properties of residual-based tests for no-cointegration applied to defactored data from which the common factors and individual components have been extracted.
    Keywords: econometrics;
    Date: 2005
  5. By: Michael Lechner; Stefan Lollivier; Thierry Magnac
    Abstract: This paper discusses the estimation of binary choice panel data models. We begin with different versions of the static random effects model when the explanatory variables are strictly exogenous. Depending on the autocorrelation structure of the errors, different estimators are available and we detail their attractiveness in each situation by trading-off their efficiency and robustness with respect to misspecification. Then, we consider the static model when a time invariant unobservable variable is correlated with the time varying explanatory variables. The non-linearity of binary choice models makes it pretty hard to eliminate individual fixed effects in likelihood functions and moment conditions, because the usual differencing out that works for the linear model does not work here except in special cases. Imposing quite restrictive assumptions is the price to pay to estimate consistently parameters of dynamics for fixed and random effects, in other words cases when the explanatory variables include lagged endogenous variables or are weakly exogenous only.
    JEL: C23
    Date: 2005–12
  6. By: Koedijk, C.G.; Tims, B.; Dijk, M.A. van (Erasmus Research Institute of Management (ERIM), RSM Erasmus University)
    Abstract: This paper analyzes the properties of multivariate tests of purchasing power parity (PPP) that fail to take heterogeneity in the speed of mean reversion across real exchange rates into account. We compare the performance of homogeneous and heterogeneous unit root testing methodologies. The recent literature has successfully contested several severe restrictions on the structure of the model, but the assumption of homogeneous mean reversion is still widely used and its consequences are virtually unexplored. Using Monte Carlo simulation, we uncover important adverse properties of the methodology that relies on homogeneous estimation and testing. More specifically, power functions are low and assume irregular shapes. Furthermore, homogeneous estimates of the mean reversion parameters exhibit potentially large biases. This can have a dramatic impact on inferences made on the validity of the PPP hypothesis. Our findings highlight the importance of allowing for heterogeneous estimation when testing for a unit root in panels of real exchange rates.
    Keywords: Purchasing Power Parity;Real Exchange Rates;Panel Models;Unit Root Tests;Heterogeneity;
    Date: 2005–12–19
  7. By: Pieralda FERRARI; Paola ANNONI
    Abstract: We propose a procedure to assess a measure for a latent phenomeno n, starting from the observation of a wide set of ordinal variabl es affected by structured missing data. The proposal is based on Nonlinear PCA technique to be jointly used with an ad hoc imputat ion method for the treatment of missing data. The procedure is pa rticularly suitable when dealing with ordinal, or mixed, variable s, which are strongly interrelated and in the presence of specifi c patterns of missing observations
    Keywords: Nonlinear PCA, monotone missing data, ordinal variables, missing data passive
  8. By: Chen Qian (Department of Economics, University of Victoria); David E. Giles (Department of Economics, University of Victoria)
    Abstract: Using small-disturbance expansions, we derive analytic expressions for the bias of the OLS estimator an elasticity in a linear model, both at an individual sample point and at the sample mean. The magnitudes of these biases are illustrated with Australian expenditure data.
    Keywords: Elasticity, bias, small-distrurbance asymptotics
    JEL: C14 C20 D12
    Date: 2005–12–22
  9. By: Johansson, Börje (CESIS - Centre of Excellence for Science and Innovation Studies, Royal Institute of Technology); Forslund, Ulla (JIBS, Jönköping International Business School)
    Abstract: Using the taxonomy by Anselin (2003), this paper investigates how the inclusion of spatially discounted variables on the ‘right-hand-side’ (RHS) in empirical spatial models affects the extent of spatial autocorrelation. The basic proposition is that the inclusion of inputs external to the spatial observation in question as a separate variable reveals spatial dependence via the parameter estimate. One of the advantages of this method is that it allows for a direct interpretation. The paper also tests to what extent significance of the estimated parameters of the spatially discounted explanatory variables can be interpreted as evidence of spatial dependence. Additionally, the paper advocates the use of the accessibility concept for spatial weights. Accessibility is related to spatial interaction theory and can be motivated theoretically by adhering to the preference structure in random choice theory. Monte Carlo Simulations show that the coefficient estimates of the accessibility variables are significantly different from zero in the case of modelled effects. The rejection frequency of the three typical tests (Moran’s I, LM-lag and LM-err) is significantly reduced when these additional variables are included in the model. When the coefficient estimates of the accessibility variables are statistically significant, it suggests that problems of spatial autocorrelation are significantly reduced. Significance of the accessibility variables can be interpreted as spatial dependence
    Keywords: accessibility; spatial dependence; spatial econometrics; Monte Carlo Simulations; spatial spillovers
    JEL: C31 C51 R15
    Date: 2005–12–28
  10. By: Juan Francisco Rubio-Ramírez; Daniel Waggoner; Tao Zha
    Abstract: This paper develops a new and easily implementable necessary and sufficient condition for the exact identification of a Markov-switching structural vector autoregression (SVAR) model. The theorem applies to models with both linear and some nonlinear restrictions on the structural parameters. We also derive efficient MCMC algorithms to implement sign and long-run restrictions in Markov-switching SVARs. Using our methods, four well-known identification schemes are used to study whether monetary policy has changed in the euro area since the introduction of the European Monetary Union. We find that models restricted to only time-varying shock variances dominate the other models. We find a persistent post-1993 regime that is associated with low volatility of shocks to output, prices, and interest rates. Finally, the output effects of monetary policy shocks are small and uncertain across regimes and models. These results are robust to the four identification schemes studied in this paper.
    Date: 2005
  11. By: Thomas H. Klier; Dan McMillen
    Abstract: A linearized version of Pinkse and Slade’s (1998) spatial probit estimator is used to account for the tendency of auto supplier plants to cluster together. By reducing estimation to two steps – standard probit or logit followed by two-stage least squares – linearization produces a model that can be estimated using large datasets. Our results imply significant clustering among older plants. Supplier plants are more likely to be in counties that are near assembly plants, that include interstate highways, and that are near other counties with supplier plants. New plants show no additional tendency toward clustering beyond that shown by older plants.
    Date: 2005

This nep-ecm issue is ©2006 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.