Heterogeneous Panel Data Models Research Articles

Determinants of Carbon Dioxide Emissions: New Empirical Evidence from MENA Countries

The paper emplyos the Pooled Mean Group (PMG) estimation to investigate empirically the relationship among carbon dioxide (CO2) emissions and potential determinants for a panel of 18 MENA countries during the period 1980-2018. The good properties in terms of consistency and efficiency of the coefficient estimates make the PMG approach very useful for examining the determinants of pollution emissions in the framework of dynamic heterogeneous panel data models over both the long- and short-run. Unlike the extant literature on MENA economies, many determinants are included in the analysis to avoid the bias problem of omitted variables. Three energy sources and two classes of sub-panels according to regional proximities and oil wealth are considered in order to provide a sensitivity check on the findings and to make the analysis more homogeneous. The results reveal long-run relationships between pollution emissions and the selected variables. All determinants are found to be statistically significant for all panels and energy sources over the short-run. However, some variables are not significant determinants over the long-run. The Environment Kuznet's Curve (EKC) hypothesis is supported only for the panel of non-oil countries, which has meaningful implications and reveals the importance of splitting the global panel in order to appropriately examine the EKC hypothesis and conduct policy debates according to the findings of each panel. The obtained results provide important policy implications.

PDMIF: Heterogeneous panel data models with interactive effects in R

PDMIF: Heterogeneous panel data models with interactive effects in R

Identification and Estimation of Dynamic Heterogeneous Unbalanced Panel Data Models with Clustering

Identification and Estimation of Dynamic Heterogeneous Unbalanced Panel Data Models with Clustering

A RMT-based LM test for error cross-sectional independence in large heterogeneous panel data models*

This paper introduces a new test for error cross-sectional independence in large panel data models with exogenous regressors having heterogenous slope coefficients. The proposed statistic, LMRMT , is based on the Lagrange Multiplier (LM) principle and the sample correlation matrix of the model’s residuals. Since in large panels poorly estimates its population counterpart, results from Random Matrix Theory (RMT) are used to establish the high-dimensional limiting distribution of LMRMT under heteroskedastic normal errors and assuming that both the panel size N and the sample size grow to infinity in comparable magnitude. Simulation results show that is largely correctly sized (except for some small values of N and T). Further, the empirical size and power outcomes show robustness of our statistic to deviations from the assumptions of normality for the error terms and of strict exogeneity for the regressors. The test has comparable small sample properties to related tests in the literature which have been developed under different asymptotic theory.

Association between economic growth, mortality and healthcare spending in 31 high-income countries

Abstract Background During the last decades, countries' healthcare spending has grown continuously, becoming a major expenditure item and threatening the overall sustainability of healthcare systems. Plenty of literature shows conflicting evidence about the relationship between economic variations and measurable health outcomes. This study aims to investigate the association between GDP, population mortality and healthcare spending in OECD high-income countries. Methods We conducted a cross-sectional study using panel data across 31 high-income countries from 2000 to 2017. Socioeconomic data for every year and each country were extracted from WHO and OECD Database. The association between current healthcare expenditure (CHE), GDP and mortality rate (MR) was investigated through a random-effects model. To control for possible reverse causality, we adopted a test of Granger causality for heterogeneous panel data models. Results The results of the random-effects model show that the MR has no statistically significant effect on CHE. We found no statistically significant association between countries' MR and GDP when the latter is the dependent variable. Our results show that an increase in GDP is associated with a significant increase of CHE (b = 0.066, p < 0.001) and CHE is significantly associated with an increase in GDP (b = 3.188, p < 0.001). The Granger causality analysis shows a unidirectional association between MR and CHE, with MR influencing CHE, albeit with a small statistical significance (p = 0.045). Between GDP and CHE, the causality is bidirectional, while between GDP and MR we found no causality. Conclusions In this study, we found a strong two-way relationship between GDP and CHE, both in the causality analysis and in the random-effect panel model. Our analysis highlights the economic multiplier effect of CHE. In the debate on the optimal allocation of resources often resulting from economic crises, this evidence should be taken into due consideration. Key messages Policymakers worldwide need to recognize the economic impact of healthcare spending when allocating financial resources. Spending on health leads to economic growth. In light of the current health-economy dichotomy, it is important to produce robust scientific evidence supporting healthcare spending.

Estimation and inference in heterogeneous spatial panels with a multifactor error structure

We develop a unifying econometric framework for the analysis of heterogeneous panel data models that can account for both spatial dependence and common factors. To tackle the challenging issues of endogeneity due to the spatial lagged term and the correlation between the regressors and factors, we propose the CCEX-IV estimation procedure that approximates factors by the cross-section averages of regressors and deals with the spatial endogeneity using the internal instrumental variables. We develop the individual and Mean Group estimators, and establish their consistency and asymptotic normality. By contrast, the Pooled estimator is shown to be inconsistent in the presence of parameter heterogeneity. Monte Carlo simulations confirm that the finite sample performance of the proposed estimators is quite satisfactory. We demonstrate the usefulness of our approach with an application to the house price growth for Local Authority Districts in the UK over 1997Q1–2016Q4.

A Lagrange-Multiplier Test for Large Heterogeneous Panel Data Models

This paper introduces a new test for error cross-sectional independence in large panel data models with exogenous regressors having heterogenous slope coefficients. The proposed statistic, LM_{RMT}, is based on the Lagrange Multiplier (LM) principle and the sample correlation matrix R_{N} of the model's residuals. Since in large panels R_{N} poorly estimates its population counterpart, results from Random Matrix Theory are used to establish the high-dimensional limiting distribution of LM_{RMT} under heteroskedastic normal errors and assuming that both the panel size N and the sample size T grow to infinity in comparable magnitude. Simulation results support our theoretical findings, with LM_{RMT} being correctly sized (except for some small values of N and T). Further, the small sample size and power outcomes show robustness of our statistic to deviations from the assumptions of normality for the error terms and regressors, of strict exogeneity for the regressors, as well as of heterogeneity for their slope coefficients. The test has comparable small sample properties to related tests in the literature which have been developed under different asymptotic theory.

A Re-examination of the Validity of the Life Cycle Hypothesis (LCH): Evidence from Emerging Europe

The study aims to empirically explore the dependence of savings behavior on demographic changes in the context of the life cycle hypothesis (LCH) in a sample of 18 European transition and post transition countries. The empirical methodology is based on a multifactor modeling approach. The research estimates heterogeneous panel data models by employing three different heterogeneous coefficient estimators: mean group (MG) estimator, common correlated effects mean group (CCEMG) estimator, and augmented mean group (AMG) estimator. The findings demonstrate that the LCH is confirmed in the case of European post-transition countries and rejected as inappropriate in European transition countries due to inconsistency of regression coefficients (age dependency, unemployment rate, urbanization, and health expenditure). The models and their findings presented in this study can be used in policymaking to predict dynamic interactions and variations among selected demographic variables in the determination of savings behavior.

Investigation on key contributors of energy consumption in dynamic heterogeneous panel data (DHPD) model for African countries: fresh evidence from dynamic common correlated effect (DCCE) approach.

The main aim of this current study is to empirically scrutinize the determinants of energy consumption for 24 African countries sub-grouped into three panels based on income levels: low-, lower-middle-, and upper-middle-income countries, from 1990 to 2015. Due to the presence of heterogeneity and cross-sectional reliance among country groups, recently developed econometric approaches, which include cross-sectional Im, Pesaran, and Shin together with cross-sectional Augmented Dickey-Fuller stationarity tests, Pedroni and Westerlund-Edgerton cointegration assessment, dynamic common correlated effect estimation approach and Dumitrescu-Hurlin Granger causality test are employed. Empirically, our findings depict analyzed variables are stationary and characterized by long-term stability affiliations for all panels. Economic growth, urbanization, population growth, and oil price with labor and capital stock as intermittent variables had palpable significant positive sway on energy consumption for all panels though their respective weight of contribution differed from one country group to another. The granger test of causation unveiled that (i) among all panels, urbanization and energy consumption are connected bidirectionally, whereas population growth causes energy consumption; (ii) a one-way causal link from economic growth to energy use is evidenced in low-income African countries, whereas a two-sided connection is confirmed in both lower-middle- and upper-middle-income economies; (iii) a bilateral causal association in low-income African nations is observed amid oil price and energy use, while a uni-lateral relationship extends from oil price to energy consumption in both lower-middle- and upper-middle-income nations in Africa. Such new methodologies and findings reveal that the long-term estimated effects as well as causal affiliations amid variables are skewed by different income levels of African countries in an attempt to conserve energy. Policy recommendations are further propose.

Heterogeneous panel data models with cross-sectional dependence

This paper considers a semiparametric panel data model with heterogeneous coefficients and individual-specific trending functions, where the random errors are assumed to be serially correlated and cross-sectionally dependent. We propose mean group estimators for the coefficients and trending functions involved in the model. It can be shown that the proposed estimators can achieve an asymptotic consistency with rates of root−NT and root−NTh, respectively as (N,T)→(∞,∞), where N is allowed to increase faster than T. Furthermore, a statistic for testing homogeneous coefficients is constructed based on the difference between the mean group estimator and a pooled estimator. Its asymptotic distributions are established under both the null and a sequence of local alternatives, even if the difference between these estimators vanishes considerably fast (can achieve root-NT2 rate at most under the null) and no consistent estimator for the covariance matrix is required. The finite sample performance of the proposed estimators together with the size and local power properties of the test are demonstrated by simulated data examples, and an empirical application with the OECD health care expenditure dataset is also provided.

Testing for causality between FDI and economic growth using heterogeneous panel data

The causal relationship between FDI inflows and growth is of great policy interest, yet the state of concrete knowledge on the issue is rather poor. Our contribution is to investigate the causal relationship between the ratio of FDI to GDP (FDIG) and economic growth (GDPG) using a battery of cutting-edge methods and an extensive data set. We employ the heterogeneous-panel tests of the Granger non-causality hypothesis based on the works of Hurlin, C. 2004a. Testing Granger Causality in Heterogeneous Panel Data Models with Fixed Coefficients. Mimeo: University of Orléans, (Fisher, R. A. 1932. Statistical Methods for Research Workers. Edinburgh: Oliver & Boyd., Fisher, R. A. 1948. ‘Combining Independent Tests of Significance.’ American Statistician 2 (5): 30–31) and Hanck, C. 2013. ‘An intersection test for panel unit roots.’ Econometric Reviews 32 (2): 183–203. Our panel data set is compiled from 136 developed and developing countries over the 1970-2006 period. According to the Hurlin and Fisher tests, FDIG unambiguously Granger-causes GDPG for at least one country. However, the results from these tests are ambiguous regarding whether GDPG Granger-causes FDIG for at least one country. Using a test based upon Hanck, C. 2013. ‘An intersection test for panel unit roots.’ Econometric Reviews 32 (2): 183–203, both with and without one structural break in the vector autoregression, we are able to determine whether and for which countries there is Granger-causality. This test suggests that at most there are six countries (Estonia, Guyana, Poland, Switzerland, Tajikistan and Yemen) where FDIG Granger-causes GDPG and at most four countries (Dominican Republic, Gabon, Madagascar and Poland) where GDPG Granger-causes FDIG.

The effect of fiscal deficits on the external imbalances in the European Union

This paper analyzes the effect of the fiscal deficit on the current account deficit in the European Union during the period 1995-2018. The purpose is to examine to what extent an increase in government spending affects the deterioration of terms of trade and contributes to increasing external imbalances. Econometric methods for heterogeneous panel data models are used to analyse the existence of a long-run relationship between the fiscal deficit and the current account. The empirical findings indicate that the twin deficits hypothesis is not confirmed for the whole European Union, but only for a certain number of member states, where a long-run relationship still exists, confirming the impact of the fiscal deficit on the current account.

Estimating latent group structure in time-varying coefficient panel data models

Summary This paper studies the estimation of latent group structures in heterogeneous time-varying coefficient panel data models. While allowing the coefficient functions to vary over cross-sections provides a good way to model cross-sectional heterogeneity, it reduces the degree of freedom and leads to poor estimation accuracy when the time-series length is short. On the other hand, in a lot of empirical studies, it is not uncommon to find that heterogeneous coefficients exhibit group structures where coefficients belonging to the same group are similar or identical. This paper aims to provide an easy and straightforward approach for estimating the underlying latent groups. This approach is based on the hierarchical agglomerative clustering (HAC) of kernel estimates of the heterogeneous time-varying coefficients when the number of groups is known. We establish the consistency of this clustering method and also propose a generalised information criterion for estimating the number of groups when it is unknown. Simulation studies are carried out to examine the finite-sample properties of the proposed clustering method as well as the post-clustering estimation of the group-specific time-varying coefficients. The simulation results show that our methods give comparable performance to the penalised-sieve-estimation-based classifier-LASSO approach by Su et al. (2018), but are computationally easier. An application to a panel study of economic growth is also provided.

Evaluating the energy-environment efficiency and its determinants in Guangdong using a slack-based measure with environmental undesirable outputs and panel data model

Environmental sustainability has become a significant goal for policymakers and practitioners since increasing environmental degradation owing to anthropogenic activities. Energy-environment efficiency, linked to a progressive reduction in the environmental impacts that may occur throughout their life cycle to levels that should be below or equal the Earth's estimated carrying capacity, is a crucial point for constructing an environment friendly society while maintaining rapid economic growth. Thus, this study combined a slack-based measure (SBM) with environmental impacts as undesirable outputs with spatial analysis techniques to measure energy-environment efficiency of 21 cities in Guangdong and its changing patterns during the period 2006–2016. What and how socioeconomic factors affecting energy-environment efficiency over time and space was further examined using heterogeneous panel data model. Here are the main findings: during the study period, energy-environment efficiency showed apparent spatiotemporal diversity with high values predominantly concentrated in coastal areas, especially in the center area of the Pearl River Delta. Energy-environment efficiency increased continuously in the western Guangdong and the Pearl River Delta, while it of eastern Guangdong showed a decreasing trend and of northern Guangdong remained stable at a low level. The results of the heterogeneous panel data model revealed that technological progress exerted the greatest positive effects on energy-environment efficiency, followed by population density, economic growth. Conversely, Openness was evaluated as an inhibiting factor. Interestingly, this study found that industrial structure demonstrated significant negative correlations with respect to energy-environment efficiency in the Pearl River Delta while it exerted significant positive influence in the peripheral areas of Guangdong. And foreign trade and energy-environment efficiency had a significant positive correlation in the Pearl River Delta, unlike the negative correlation in the peripheral areas of Guangdong. This study's findings hold a helpful reference for both policymakers and practitioners to coordinate the economy, energy and environment and established environment-friendly society in the fast-developed areas like Guangdong.

Does E-government reduce corruption? Evidence from a heterogeneous panel data model

PurposeThe purpose of this paper is to revisit the relationship between E-government and corruption using global panel data from 176 countries covering the period from 2003 to 2014, considering other potential determinants, such as economic prosperity (gross domestic product per capita [GDPPC]), price stability (inflation), good governance (political stability and government effectiveness) and press freedom (civil liberties and political rights) indicators. Hence, the main rationale of this study is to reexamine the conventional wisdom as to the relationship between E-government and corruption using panel data independent of any preexisting notions.Design/methodology/approachThe probability reduction approach of empirical modeling proposed by Spanos (2009) is used to test the relationship. Secondary data were collected from the United Nations, the World Bank, Transparency International and Freedom House.FindingsNo statistical evidence was found for the idea that E-government has a positive impact on corruption reduction following a rigorous test of the proposition. However, strong evidence was found for the positive impact of a country’s government effectiveness, political stability and economic status. There also appears to be some evidence for the effect of GDPPC and civil liberties. There is no evidence to prove that inflation and political rights have any corruption reducing the effect.Research limitations/implicationsCase studies suggest that E-government is helpful for curbing corruption. This study includes and examines some of the potential and important variables associated with corruption. Further research is encouraged and it should include more variables, such as national culture, poverty, religion and geography. Regarding methodology, a more parsimonious model must be sought to take into account adequately the entire probabilistic structure of the data.Practical implicationsThe findings of the study demonstrate that E-government is less significant for reducing corruption compared to other factors. Hence, policymakers should further focus on other potential areas such as socio-economic factors, good governance, culture and transparency to combat corruption in addition to improving digital government.Originality/valueThis research applies a new methodological approach to the study of the relationship between E-government and corruption.

Drivers of greenhouse gas emissions in the United States: revisiting STIRPAT model

The challenge of reducing emissions of greenhouse gases (GHG) has stimulated great attention among policymakers and scholars in recent past, and a number of STIRPAT (Stochastic Impacts by Regression on Population, Affluence, and Technology) studies on carbon emissions have been conducted. This paper contributes to that literature by: (i) studying per capita GHG emissions in the United States (US) adopting STIRPAT modeling framework; (ii) employing new explanatory factors like cattle population density, political willingness to address environmental problems, and educational attainment; and (iii) investigating whether emissions elasticities of various factors vary within the US or not. State-level panel data over the period 1990–2014 are used, and partitioning of the sample is done with respect to two controlling factors: an indicator of political support to environmentalism and educational attainment. Results of heterogeneous slope parameters panel data models indicate that cattle density and affluence are major drivers of per capita GHG emissions in the continental US. We find strong evidence of heterogeneity in emissions elasticities across partitioned samples. Our grouping analysis suggests that in a diverse country like US, policymakers should not focus on the average relationships dictated by a single STIRPAT equation, but should account for regional differences if they want accuracy and higher effectiveness in climate policymaking.

Sieve Estimation of Time-Varying Panel Data Models With Latent Structures

We propose a heterogeneous time-varying panel data model with a latent group structure that allows the coefficients to vary over both individuals and time. We assume that the coefficients change smoothly over time and form different unobserved groups. When treated as smooth functions of time, the individual functional coefficients are heterogeneous across groups but homogeneous within a group. We propose a penalized-sieve-estimation-based classifier-Lasso (C-Lasso) procedure to identify the individuals’ membership and to estimate the group-specific functional coefficients in a single step. The classification exhibits the desirable property of uniform consistency. The C-Lasso estimators and their post-Lasso versions achieve the oracle property so that the group-specific functional coefficients can be estimated as well as if the individuals’ membership were known. Several extensions are discussed. Simulations demonstrate excellent finite sample performance of the approach in both classification and estimation. We apply our method to study the heterogeneous trending behavior of GDP per capita across 91 countries for the period 1960–2012 and find four latent groups.

A two-stage estimator for heterogeneous panel models with common factors

The estimation in a stationary heterogeneous panel model where unknown common factors are present is considered. A two-stage estimator is proposed and compared to existing alternative methods for the estimation of slope parameters in panels with a multifactor error structure. The asymptotic properties of this estimator are provided alongside the comparison of its finite-sample properties with those of a range of estimators by means of Monte Carlo experiments. The two-stage estimator affords a greater computational simplicity with respect to existing iterative estimators that fail to achieve convergence in a relevant number of cases considered. The proposed estimator is employed in the analysis of the determinants of Euro-denominated bonds in a framework of a heterogeneous panel data model with multifactor error structure.

Heterogeneous Panel Data Models with Cross--Sectional Dependence

This paper considers a semiparametric panel data model with heterogeneous coefficients and individual-specific trending functions, where the random errors are assumed to be serially correlated and cross-sectionally dependent. We propose mean group estimators for the coefficients and trending functions involved in the model. It can be shown that the proposed estimators can achieve an asymptotic consistency with rates of root-NT and root-NTh, respectively, where N is allowed to increase faster than T. Furthermore, a statistic for testing homogeneous coefficients is constructed based on the difference between the mean group estimator and a pooled estimator. Its asymptotic distributions are established under both the null and a sequence of local alternatives, even if the difference between these estimators vanishes considerably fast (can achieve root-NT rate at most under the null) and consistent estimator available for the covariance matrix is not required explicitly. The finite sample performance of the proposed estimators together with the size and local power properties of the test are demonstrated by simulated data examples, and an empirical application with the OECD health care expenditure dataset is also provided.

A Bayesian heterogeneous coefficients spatial autoregressive panel data model of retail fuel duopoly pricing

We apply a heterogenous coefficient spatial autoregressive panel model to explore competition/cooperation by duopoly pairs of German fueling stations in setting prices for diesel and e5 fuel. We rely on a Markov Chain Monte Carlo (MCMC) estimation methodology applied with non-informative priors, which produces estimates equivalent to those from (quasi-) maximum likelihood. We explore station-level pricing behavior using pairs of proximately situated fueling stations with no nearby neighbors. Our sample data represents average daily diesel and e5 fuel prices, and refinery cost information covering more than 487 days.The heterogeneous coefficients spatial autoregressive panel data model uses the large sample of daily time periods to produce spatial autoregressive model estimates for each fueling station. These estimates provide information regarding the price reaction function of each station to its duopoly rival station. This is in contrast to conventional estimates of price reaction functions that average over the entire cross-sectional sample of stations. We show how these estimates can be used to infer competition versus cooperation in price setting by individual stations.