762 research outputs found
Weak and Strong Cross-Sectional Dependence: A Panel Data Analysis of International Technology Diffusion
No full textInternational audienc
Interaction matrix selection in spatial autoregressive models with an application to growth theory
No full textInternational audienceThe interaction matrix, or spatial weight matrix, is the fundamental tool to model cross-sectional interdependence between observations in spatial autoregressive models. However, it is most of the time not derived from theory, as it should be ideally, but chosen on an ad hoc basis. In this paper, we propose a modified version of the J test to formally select the interaction matrix. Our methodology is based on the application of the robust against unknown heteroskedasticity GMM estimation method, developed by Lin and Lee (2010). We then implement the testing procedure developed by Hagemann (2012) to overcome the decision problem inherent to non-nested models tests. An application of the testing procedure is presented for the Schumpeterian growth model with worldwide interactions developed by Ertur and Koch (2011) using three different types of interaction matrices: genealogic distance, linguistic distance and bilateral trade flows. We find that the interaction matrix based on trade flows is the most adequate
The European Enlargement Process and Regional Convergence Revisited: Spatial Effects Still Matter.
No full textThis paper has two main goals. First, it reconsiders regional growth and convergence processes in the context of the enlargement of the European Union to new member states. We show that spatial autocorrelation and heterogeneity still matter in a sample of 237 regions over the period 1993-2002. Spatial convergence clubs are defined using exploratory spatial data analysis and a spatial autoregressive model is estimated. We find strong evidence that the growth rate of per capita GDP for a given region is positively affected by the growth rate of neighbouring regions. The second objective is to test the robustness of the results with respect to non-normality, outliers and heteroskedasticity using two other methods: The quasi maximum Likelihood and the Bayesian estimation methods.
Growth and Spatial Dependence - The Mankiw, Romer and Weil model revisited
No full textThe aim of this paper is to analyze the theoretical and econometric implications of omitting spatial dependence in the Mankiw, Romer, and Weil model. Indeed, the international distribution of income levels and growth rates suggests the existence of large international disparities, and therefore the important role of location on economic performance. However, taking spatial dependence into account requires resorting to the methods of Spatial Econometrics, not only for a valid statistical inference, but also for revaluating the impact of the variables generally considered as crucial in the growth phenomenon and finding the processes underlying growth rates and income levels.
Learning Complex Policy Distribution with CEM Guided Adversarial Hypernetwork
No full textCross-Entropy Method (CEM) is a gradient-free direct policy search method, which has greater stability and is insensitive to hyperparameter tuning. CEM bears similarity to population-based evolutionary methods, but, rather than using a population it uses a distribution over candidate solutions (policies in our case). Usually, a natural exponential family distribution such as multivariate Gaussian is used to parameterize the policy distribution. Using a multivariate Gaussian limits the quality of CEM policies as the search becomes confined to a less representative subspace. We address this drawback by using an adversarially-trained hypernetwork, enabling a richer and complex representation of the policy distribution. To achieve better training stability and faster convergence, we use a multivariate Gaussian CEM policy to guide our adversarial training process. Experiments demonstrate that our approach outperforms state-of-the-art CEM-based methods by 15.8% in terms of rewards while achieving faster convergence. Results also show that our approach is less sensitive to hyper-parameters than other deep-RL methods such as REINFORCE, DDPG and DQN.Interactive Intelligenc
"Dual' gravity: Using spatial econometrics to control for multilateral resistance"
No full textWe propose a quantity-based 'dual' version of the gravity equation that yields an estimating equation with both cross-sectional interdependence and spatially lagged error terms. Such an equation can be concisely estimated using spatial econometric techniques. We illustrate this methodology by applying it to the Canada-U.S. data set used previously, among others, by Anderson and van Wincoop (2003) and Feenstra (2002, 2004). Our key result is to show that controlling directly for spatial interdependence across trade flows, as suggested by theory, significantly reduces border effects because it captures 'multilateral resistance'. Using a spatial autoregressive moving average specification, we find that border effects between the U.S. and Canada are smaller than in previous studies: about 8 for Canadian provinces and about 1.3 for U.S. states. Yet, heterogeneous coefficient estimations reveal that there is much variation across provinces and states.
The European Regional Convergence Process, 1980-1995: Do Spatial Regimes and Spatial Dependence Matter?
No full textWe show in this paper that spatial dependence and spatial heterogeneity matter in the estimation of the b-convergence process among 138 European regions over the 1980-1995 period. Using spatial econometrics tools, we detect both spatial dependence and spatial heterogeneity in the form of structural instability across spatial convergence clubs. The estimation of the appropriate spatial regimes spatial error model shows that the convergence process is different across regimes. We also estimate a strongly significant spatial spillover effect: the average growth rate of per capita GDP of a given region is positively affected by the average growth rate of neighboring regions.convergence, club convergence, spatial econometrics, European regions, spatial regimes, spatial autocorrelation
‘Dual’ gravity: using spatial econometrics to control for multilateral resistance
No full textWe propose a quantity-based `dual' version of the gravity equation that yields an estimating equation with both cross-sectional interdependence and spatially lagged error terms. Such an equation can be concisely estimated using spatial econometric techniques. We illustrate this methodology by applying it to the Canada-U.S. data set used previously, among others, by Anderson and van Wincoop (2003) and Feenstra (2002, 2004). Our key result is to show that controlling directly for spatial interdependence across trade flows, as suggested by theory, significantly reduces border effects because it captures `multilateral resistance'. Using a spatial autoregressive moving average specification, we find that border effects between the U.S. and Canada are smaller than in previous studies: about 8 for Canadian provinces and about 1.3 for U.S. states. Yet, heterogeneous coefficient estimations reveal that there is much variation across provinces and states.gravity equations, multi-region general equilibrium trade models; spatial econometrics, border effects
Tests de non stationnarité et tendances non linéaires
No full textThis paper stresses the importance of the hypothesis of linearity of the deterministic component imposed by unit root testing procedures most frequently used in empirical literature. We suggest an empirical testing strategy which reduces the risk of reaching false conclusions due to the misspecification of that component and we apply it to the analysis of the nonstationarity exhibited by real GNP in France. We show that it is possible to find someflexible specifications which enable us to reject the unit root null hypothesis otherwise strongly supported in empirical literature. These specifications might be considered as approximations of the true process generating real GNP and might be useful for other time series as well.Nous nous proposons dans ce travail d’attirer l ’attention sur le postulat de linéarité de la composante déterministe imposé par les procédures de test de la racine unitaire les plus utilisées dans la littérature empirique. Nous avons tenté de mettre au point une stratégie empirique permettant de réduire le risque de parvenir à des conclusions erronées suite à une mauvaise spécification de la composante déterministe et nous l ’avons appliquée à l’étude de la non stationnarité caractérisant le PIB réel marchand CVS en France. Nous avons ainsi mis en évidence des spécifications flexibles permettant de rejeter l’hypothèse nulle de la racine unitaire fortement étayée dans la littérature empirique. Ces spécifications peuvent être considérées comme des approximations du vrai processus engendrant la série du PIB réel mais peuvent également se révéler utiles dans l ’étude d’autres séries chronologiques
Tests de non stationnarité et tendances non linéaires
No full textThis paper stresses the importance of the hypothesis of linearity of the deterministic component imposed by unit root testing procedures most frequently used in empirical literature. We suggest an empirical testing strategy which reduces the risk of reaching false conclusions due to the misspecification of that component and we apply it to the analysis of the nonstationarity exhibited by real GNP in France. We show that it is possible to find someflexible specifications which enable us to reject the unit root null hypothesis otherwise strongly supported in empirical literature. These specifications might be considered as approximations of the true process generating real GNP and might be useful for other time series as well.Nous nous proposons dans ce travail d’attirer l ’attention sur le postulat de linéarité de la composante déterministe imposé par les procédures de test de la racine unitaire les plus utilisées dans la littérature empirique. Nous avons tenté de mettre au point une stratégie empirique permettant de réduire le risque de parvenir à des conclusions erronées suite à une mauvaise spécification de la composante déterministe et nous l ’avons appliquée à l’étude de la non stationnarité caractérisant le PIB réel marchand CVS en France. Nous avons ainsi mis en évidence des spécifications flexibles permettant de rejeter l’hypothèse nulle de la racine unitaire fortement étayée dans la littérature empirique. Ces spécifications peuvent être considérées comme des approximations du vrai processus engendrant la série du PIB réel mais peuvent également se révéler utiles dans l ’étude d’autres séries chronologiques
- …
