Heterogeneous Panel Data Models Research Articles

A Lagrange Multiplier test for cross-sectional dependence in a fixed effects panel data model

It is well known that the standard Breusch and Pagan (1980) LM test for cross-equation correlation in a SUR model is not appropriate for testing cross-sectional dependence in panel data models when the number of cross-sectional units (n) is large and the number of time periods (T) is small. In fact, a scaled version of this LM test was proposed by Pesaran (2004) and its finite sample bias was corrected by Pesaran et al. (2008). This was done in the context of a heterogeneous panel data model. This paper derives the asymptotic bias of this scaled version of the LM test in the context of a fixed effects homogeneous panel data model. This asymptotic bias is found to be a constant related to n and T, which suggests a simple bias corrected LM test for the null hypothesis. Additionally, the paper carries out some Monte Carlo experiments to compare the finite sample properties of this proposed test with existing tests for cross-sectional dependence.

Testing for Granger non-causality in heterogeneous panels

This paper proposes a very simple test of Granger (1969) non-causality for heterogeneous panel data models. Our test statistic is based on the individual Wald statistics of Granger non causality averaged across the cross-section units. First, this statistic is shown to converge sequentially to a standard normal distribution. Second, the semi-asymptotic distribution of the average statistic is characterized for a fixed T sample. A standardized statistic based on an approximation of the moments of Wald statistics is hence proposed. Third, Monte Carlo experiments show that our standardized panel statistics have very good small sample properties, even in the presence of cross-sectional dependence.

A dynamic panel data analysis for relationship between private and public investment in R and D

This paper investigates the relationship between private and public investment in R and D, while taking into account the effect of several instruments policies such as subsidies and taxes. We design a new look of knowledge spillovers and R and D cooperation to explain the contribution of public and private R and D on growth. We propose a heterogeneous dynamic panel data model to consider the effect as well as endogenous. We also distinguish between the estimated long run and short run results. Our results based on a sample of 23 countries over the period 1992 - 2009 indicate that both private and private investment in R and D are complement. By establishing an endogenous growth model, the estimates indicate that public and private R and D depend on the host country’s human capital investment and that FDI is a more significant spillover channel than imports.

Seemingly unrelated regressions with spatial error components

This article considers various estimators using panel data seemingly unrelated regressions (SUR) with spatial error correlation. The true data generating process (DGP) is assumed to be SUR with spatial error of the autoregressive or moving average type. Moreover, the remainder term of the spatial process is assumed to follow an error component structure. Both maximum likelihood (ML) and generalized moments (GM) methods of estimation are used. Using Monte Carlo experiments, we check the performance of these estimators and their forecasts under misspecification of the spatial error process, various spatial weight matrices, and heterogeneous versus homogeneous panel data models.

Seemingly Unrelated Regressions with Spatial Error Components

This paper considers various estimators using panel data seemingly unrelated regressions (SUR) with spatial error correlation. The true data generating process is assumed to be SUR with spatial error of the autoregressive or moving average type. Moreover, the remainder term of the spatial process is assumed to follow an error component structure. Both maximum likelihood and generalized moments (GM) methods of estimation are used. Using Monte Carlo experiments, we check the performance of these estimators and their forecasts under misspecification of the spatial error process, various spatial weight matrices, and heterogeneous versus homogeneous panel data models.

Foreign Capital Inflows, Economic Policies and the Real Exchange Rate in Sub Saharan Africa: Is There an Interaction Effect?

This paper presents a dynamic heterogeneous panel data model in which the reaction of the real exchange rate to external finance includes interactions with trade openness and the fiscal, monetary and nominal exchange rate policies of twenty-four Sub-Saharan African countries from 1978-2001. As expected, a rise in international transfers exerts a real appreciation pressure in reform-compliant economies. However, the results from this paper indicate that this estimated inflationary effect of capital inflows could be effectively contained by associated policy interventions to contract government expenditure, relax rigidities in factor markets, liberalise trade controls and address problems of credit rationing in the private sector.

Forecasting with panel data

AbstractThis paper gives a brief survey of forecasting with panel data. It begins with a simple error component regression model and surveys the best linear unbiased prediction under various assumptions of the disturbance term. This includes various ARMA models as well as spatial autoregressive models. The paper also surveys how these forecasts have been used in panel data applications, running horse races between heterogeneous and homogeneous panel data models using out‐of‐sample forecasts. Copyright © 2008 John Wiley & Sons, Ltd.

A longitudinal analysis of the impact of firm resources and industry characteristics on firm-specific profitability

Using a dynamic heterogeneous panel data model, we examine the relationship between firm-specific resources (corporate management capabilities, employee value-added and technological competence) and firm-specific profitability and the potential moderating effects of industry characteristics on this relationship. We find that firm-specific resources enhance both accounting-based measures (return on assets and return on sales) and market-based measure (Tobin’s q) of firm-specific performance. Moreover, industry characteristics moderate the relationship between firm-specific resources and firm-specific profitability. Managerial implications are discussed.

International Investment Positions and Exchange Rate Dynamics: A Dynamic Panel Analysis

In this paper we revisit medium- to long-run exchange rate determination, focusing on the role of international investment positions. To do so, we develop a new econometric framework accounting for conditional long-run homogeneity in heterogeneous dynamic panel data models. In particular, in our model the long-run relationship between effective exchange rates and domestic as well as weighted foreign prices is a homogeneous function of a country's international investment position. We find rather strong support for purchasing power parity in environments of limited negative net foreign asset to GDP positions, but not outside such environments. We thus argue that the purchasing power parity hypothesis holds conditionally, but not unconditionally, and that international investment positions are an essential component to characterizing this conditionality. Finally, we adduce evidence that whether deterioration of a country's net foreign asset to GDP position leads to a depreciation of that country's effective exchange rate depends on its rate of inflation relative to the rate of inflation abroad as well as its exposure to global shocks.

Modelling multi‐period inflation uncertainty using a panel of density forecasts

AbstractThis paper examines the determinants of inflation forecast uncertainty using a panel of density forecasts from the Survey of Professional Forecasters (SPF). Based on a dynamic heterogeneous panel data model, we find that the persistence in forecast uncertainty is much less than what the aggregate time series data would suggest. In addition, the strong link between past forecast errors and current forecast uncertainty, as often noted in the ARCH literature, is largely lost in a multi‐period context with varying forecast horizons. We propose a novel way of estimating ‘news’ and its variance using the Kullback‐Leibler information, and show that the latter is an important determinant of forecast uncertainty. Our evidence suggests a strong relationship of forecast uncertainty with level of inflation, but not with forecaster discord or with the volatility of a number of other macroeconomic indicators. Copyright © 2006 John Wiley & Sons, Ltd.

Forecasting with Panel Data

This paper gives a brief survey of forecastiang with panel data. Starting with a simple error component regression model and surveying best linear unbiased prediction under various assumptions of the disturbance term. This includes various ARMA models as well as spatial autoregressive models. The paper also surveys how these forecasts have been used in panel data applications, running horse races between heterogeneous and homogeneous panel data models using out of sample forecasts.

Forecasting with Panel Data

This paper gives a brief survey of forecasting with panel data. Starting with a simple error component regression model and surveying best linear unbiased prediction under various assumptions of the disturbance term. This includes various ARMA models as well as spatial auto-regressive models. The paper also surveys how these forecasts have been used in panel data applications, running horse races between heterogeneous and homogeneous panel data models using out of sample forecasts.

Reconsidering the Evidence

This paper, using 40 years of monthly industrial production data, examines the relationship between the business cycles of the 12 euro area countries. Since estimates of the business cycle have been found to be sensitive to how the cycle is measured, a range of alternative measures is considered. We focus on both parametric and nonparametric univariate measures of the "classical" and "growth" cycles. We then investigate whether euro area business cycles have converged. This is based on a descriptive analysis of the distribution of bivariate correlation coefficients between the 12 countries’ business cycles. This extends previous work that has looked for convergence, in a similar manner by focusing on correlation, but has not considered the entire distribution, instead focusing on the mean correlation coefficient or particular bivariate correlation coefficients. Moreover, exploiting the panel of correlation coefficients we propose a statistical test for convergence based on estimation of a dynamic heterogeneous panel data model. Although empirical inference about individual euro area business cycles is found to be sensitive to the measure of the business cycle considered, our findings about convergence between the euro area business cycles exhibit similarities across the alternative measures of the business cycle. Interestingly, we find that there have been periods of convergence, identified by the distribution tending to unity, and periods of divergence. The most recent estimates suggest that correlation between the 12 European cycles is statistically positive, and has risen from a trough in the early 1990s. This is confirmed by the test for convergence, which indicates that, despite some volatility over the last 20 years, the long-run trend is for rising correlation between euro area business cycles.

Un test simple de l'hypothèse de non-causalité dans un modèle de panel hétérogène

Cet article propose un test simple de l’hypothese de non-causalite au sens de Granger [1969] dans un modele de panel heterogene avec coefficients fixes. Nous proposons une statistique de test fondee sur la moyenne des statistiques de Wald associees aux tests individuels de non-causalite. Cette statistique converge sequentiellement vers une distribution normale. Pour une taille T fixee, une seconde statistique standardisee utilisant une approximation des moments des statistiques de Wald individuelles est proposee. Des simulations montrent que ces deux statistiques possedent de bonnes proprietes sur des echantillons de faible dimension temporelle.