Linear Panel Models Research Articles

Fundamental pricing laws and long memory effects in the day-ahead power market

We apply linear and nonlinear panel models to investigate fundamental pricing laws in the Greek day-ahead electricity market. We link the persistence behaviour and other dynamic properties of electricity price trajectories to an extended array of fundamental variables reflecting market conditions and the price course of energy commodities in global markets. Assuming that the dynamics of the intraday power pricing curve is adequately specified within a panel modelling framework with long memory and common unobserved stochastic effects, we are able to provide consistent estimates of price responsiveness to the main fundamental drivers and thus avoid biases induced by other popular modelling approaches. We also device a panel variant of the standard smooth-transition regression model to investigate regime switching effects in the price responsiveness mechanism. Empirical results suggest that Greek day-ahead electricity market prices respond significantly to hour-specific (load, marginal generation capacity) and global (Brent price, emission allowance rights cost) fundamental variables. In some intraday power delivery zones, these covariates are in a fractional cointegrating relationship with electricity prices being responsible for their drifting behaviour. The strength and type of cointegration changes across intraday trading periods in response to market conditions. The results of the nonlinear model demonstrate the ineffectiveness of linear model designs (either univariate or panel) to accurately estimate the levels of price risk and associate it with changes in the market climate.

Social origins, geographical mobility and occupational attainment in contemporary Italy

This paper studies the effect in the Italian case of geographical mobility on employment and occupational attainment, defined as access to the upper class, avoidance of the working class, and avoidance of agricultural jobs. It observes the distribution of its effect over the life course. Given that migration is a gendered phenomenon, we perform separate analyses by gender. Our data set, moreover, includes residential information at the municipality level, making it possible to specify geographical mobility in different ways, according to the distance, the characteristics of origin, and destination and the frequency of individual movements. Third, it studies whether the effects of geographical mobility change according to social class of origin and geographical area of origin.Our analyses, based on linear probability panel models with fixed effects, show a strong gender divide concerning the probability of employment and avoidance of the working class. A positive effect of geographical mobility on occupational outcomes appears to exist only as regards men, because for women the divergence between movers and stayers appears well before geographical mobility. Finally, the effects of geographical mobility are generally stronger for individuals originating from the middle and lower classes and from rural areas, but they are not so strong as to enable those individuals to substantially change their position in the occupational hierarchy.

Fundamental Pricing Laws and Long Memory Effects in the Day-Ahead Power Market

We apply linear and nonlinear panel models to investigate fundamentals pricing relationships in the Greek day-ahead power market. We link the persistence behaviour and other dynamic properties of power price trajectories to an extended array of fundamental variables reflecting market conditions and the price course of energy commodities in global markets. Assuming that the dynamics of the intraday power pricing curve is adequately specified within a panel modelling framework with long memory and common unobserved stochastic effects, we are able to provide consistent estimates of price responsiveness to the main fundamental drivers and thus avoid biases induced by other popular modelling approaches. We also device a panel variant of the standard smooth-transition regression model to investigate regime switching effects in the price responsiveness mechanism. Empirical results support that Greek day-ahead prices are in a fractional cointegrating relationship with hour-specific covariates (load, marginal generation capacity) as well as price determinants affecting the entire cross-section of hourly contracts (Brent price, emission allowance rights cost). The strength and type of cointegration often changes in a nonlinear way with the hour of power delivery or in response to market conditions. Our study has important implications for price risk management. The results of the nonlinear model demonstrate the ineffectiveness of linear models (either univariate or in a panel context) to accurately estimate the levels of price risk and associate it with changes in market conditions.

Visualization, Identification, and Estimation in the Linear Panel Event-Study Design

Linear panel models, and the that often accompany them, are popular tools for learning about policy effects. We discuss the construction of event-study plots and suggest ways to make them more informative. We examine the economic content of different possible identifying assumptions. We explore the performance of the corresponding estimators in simulations, highlighting that a given estimator can perform well or poorly depending on the economic environment. An accompanying Stata package, -xtevent-, facilitates adoption of our suggestions. Institutional subscribers to the NBER working paper series, and residents of developing countries may download this paper without additional charge at www.nber.org.

Bayesian estimation and model comparison for linear dynamic panel models with missing values

SummaryPanel data are collected over several time periods for the same units and hence allow for modelling both latent heterogeneity and dynamics. Since in a dynamic setup, the dependent variable also appears as an explanatory variable in later periods, missing values lead to substantial loss of information and the possibility of inefficient estimation. For linear dynamic panel models with fixed or random effects, we suggest a Bayesian approach to deal with missing values. The Gibbs sampling scheme providing a sample from the posterior distribution is thereby augmented by draws from the full conditional distribution of the missing values. While the full conditional distribution for missing values in the dependent variable is implied by the model setup, we incorporate a flexible non‐parametric approximation to the full conditional posterior distribution of missing values in the explaining variables. Also, we provide accurate non‐nested model comparison in terms of the marginal likelihood from the resulting hybrid Gibbs sampling output. The properties and possible efficiency gains of the suggested approach are illustrated by means of a simulation study and an empirical application using a macroeconomic panel data set.

Fitting partially linear functional-coefficient panel-data models with Stata

In this article, we describe the implementation of fitting partially linear functional-coefficient panel models with fixed effects proposed by An, Hsiao, and Li [2016, Semiparametric estimation of partially linear varying coefficient panel data models in Essays in Honor of Aman Ullah ( Advances in Econometrics, Volume 36)] and Zhang and Zhou (Forthcoming, Econometric Reviews). Three new commands xtplfc, ivxtplfc, and xtdplfc are introduced and illustrated through Monte Carlo simulations to exemplify the effectiveness of these estimators.

Common Determinants of Credit Default Swap Premia in the North American Oil and Gas Industry. A Panel BMA Approach

This study discovered market determinants of credit default swap (CDS) spreads in the North American oil and gas industry. Due to the limited theoretical background on market sources of CDS price fluctuations, we chose to alleviate model uncertainty and possible misspecification issues using Bayesian model averaging. This robust framework allowed us to aggregate results from a vast number of linear panel models estimated over the 2017–2020 period. We identified oil price volatility, major shifts in the OPEC+ supply policy, natural gas prices and industrial metal prices as the most robust determinants of CDS spreads. We show that following the onset of the COVID-19 pandemic, oil prices ceased to be a notably important determinant of credit risk, as factors indirectly related to oil prices, such as global and sectoral uncertainty, financial conditions and the macroeconomic stance became more influential. Additionally, we show that the CDS spreads of shale companies are determined by similar common factors, but they are more sensitive to the OPEC+ decisions on the global supply and are less affected by the domestic activity. Finally, we also prove that our modelling approach may help investors and risk officers to identify robust determinants behind the dynamics of credit risk.

Tests for p-regression Coefficients in Linear Panel Model When p is Divergent

This paper evaluates the performance of the FW-test for testing part of p-regression coefficients in linear panel data model when p is divergent. The asymptotic power of the FW-statistic is obtained under some regular conditions. The theoretical development are challenging since the number of covariates increases as the sample size increases. It is worth noting that the inference approach does not require any specification of the error distribution. Some simulation comparisons are conducted and show that the simulated power coincide with theoretical power well. The method is also illustrated using a renal cancer data example.

Impact of Information and Communication Technology Diffusion on HIV and Tuberculosis Health Outcomes among African Health Systems

Debate regarding the impact of information and communication technology (ICT) on health outcomes has prompted researchers to conduct analyses across many parts of the globe, yet, still little is known about the ICT impact in the African continent. Using a robust multivariate approach, this study examined system-wide impact of ICT diffusion on multiple health outcomes for HIV and tuberculosis among sovereign countries of Africa. This study utilized longitudinal panel data from the World Bank and International Telecommunication Union databases between 2000 and 2016. We relied on a robust linear dynamic panel model to incorporate lagged time variables to estimate the relationships between ICT infrastructure (mobile phone use, internet access, and fixed-telephone subscriptions) and HIV and tuberculosis outcomes. Econometric analyses found that the coefficients of the aggregate ICT variables were all negative (except for fixed telephones) for tuberculosis health measures and HIV prevalence, and positive for access to antiretroviral therapy. The diffusion of mobile phones and internet was associated with decreased incidence of tuberculosis, HIV prevalence, and tuberculosis mortality rates. However, increased diffusion of these three ICT tools was associated with increased access to antiretroviral therapy. Thus, African governments should identify investment strategies for adopting and implementing ICT to improve population health outcomes.

A portmanteau test for serial correlation in a linear panel model

We introduce the command xtserialpm to perform the portmanteau test developed in Jochmans (2019, Cambridge Working Papers in Economics No. 1993, University of Cambridge, Faculty of Economics). The procedure tests for serial correlation of arbitrary form in the errors of a linear panel model after estimation of the regression coefficients by the within-group estimator. The test is designed for short panels and can deal with general missing-data patterns. The test is different from the related portmanteau test of Inoue and Solon (2006, Econometric Theory 22: 835–851), which is performed by xtistest (Wursten, 2018, Stata Journal 18: 76–100), in that it allows for heteroskedasticity. In simulations documented below, xtserialpm is found to provide a more powerful test than xthrtest (Wursten 2018), which performs the test for first-order autocorrelation of Born and Breitung (2016, Econometric Reviews 35: 1290–1316). We also provide comparisons with xtistest and xtserial (Drukker, 2003, Stata Journal 3: 168–177). These tests perform well under stationarity but break down under even mild forms of heteroskedasticity.

Does financial inclusion impact CO2 emissions? Evidence from Asia

This study examines the impact of financial inclusion on CO2 emissions using a sample of 31 Asian countries during the period 2004–2014. Three composite indicators for financial inclusion are constructed using principal component analysis (PCA) based on normalized variables. To estimate the model, we adopted the Hoechle (2007) procedure which produces Driscoll-Kraay standard errors for linear panel models that are not only heteroskedasticity consistent but also robust to general forms of cross-sectional dependence. We find that income, energy consumption, industrialization, urbanization, FDI and financial inclusion appear to have led to higher emissions of CO2 in the region. Meanwhile, increased openness to trade seems to have reduced CO2 emissions. The findings are qualitatively robust to different proxies of financial inclusions and reasonable modifications to specification of the model. The empirical results imply that there are currently no policy synergies between growing financial inclusion and mitigating CO2 emissions. Thus, financial inclusion should be integrated into climate change adaptation strategies at local, national and regional levels, especially to address the side effect of higher CO2 emissions associated with improved financial inclusion.

Can Money Buy Happiness? Subjective Wellbeing and Its Relationship With Income, Relative Income, Monetary and Non-Monetary Poverty in Bangladesh

This paper presents an empirical analysis of the importance of income, relative income, monetary and non-monetary poverty for individual wellbeing or happiness in rural Bangladesh. The study is the first estimate of a wellbeing function for Bangladesh using nationally representative micro-panel data. Employing a linear panel model with individual random effects and a large set of control variables like education, working status and disability, we found a strong and positive relationship between wellbeing and income. Being further below the poverty line – estimated using the depth-ofpoverty measure – was found to have a significant negative effect on happiness. On the other hand, the income of the reference group was found to be just as important as one’s own income for an individual’s happiness. Comparisons were found to be asymmetric and upwards. Improvements in a multidimensional poverty index, constructed using indicators of household education, health and living standards, were found to have a positive and significant relationship with happiness in all specifications. Gender-disaggregated analysis reveals that, while the income effect was found to be stronger for male individuals, the effects of relative income, monetary and nonmonetary poverty on subjective wellbeing are larger for female individuals. The results thus point towards a need to incorporate such notions into the assessment of individuals’ wellbeing.

Identification and estimation in panel models with overspecified number of groups

We propose a simple and fast approach to identify and estimate the unknown group structure in panel models by adapting the M-estimation method. We consider both linear and nonlinear panel models where the regression coefficients are heterogeneous across groups but homogeneous within a group and the group membership is unknown to researchers. The main result of the paper is that under certain assumptions, our approach is able to provide uniformly consistent estimation as long as the number of groups used in estimation is not smaller than the true number of groups. We also show that, asymptotically, our method may partition some true groups into further subgroups, but cannot mix units from different groups. When the true number of groups is used in estimation, all units can be categorized correctly with probability approaching one, and we establish the limiting distribution for the estimators of the group parameters. In addition, we provide an information criterion to select the number of groups, and establish the consistency of the selection criterion under some mild conditions. Monte Carlo simulations are conducted to examine the finite sample performance of the proposed method. The findings in the simulation confirm our theoretical results in the paper. Applications to two real datasets also highlight the necessity to consider both individual heterogeneity and group heterogeneity in the model.

Financial development and energy consumption: Is the MENA region different?

This paper examines the relationship between financial development and energy consumption estimations in the major MENA countries (Algeria, Bahrain, Egypt, Iran, Iraq, Israel, Jordan, Kuwait, Lebanon, Libya, Morocco, Oman, Qatar, Saudi Arabia, Syria, Tunisia, United Arab Emirates and Yemen) over the period 1996 to 2014. We consider the energy use in kg of oil equivalent per capita as a dependent variable that reflects the cross-country energy consumption. We measure the level of financial development in the MENA countries by considering banking indicators. Extending the model of Sadorsky (2011), we estimate both linear and non-linear dynamic panel model. We use new robust econometrics to take into account heterogeneity and nonlinearity and we control the estimation results for the period of the global financial crisis. The results of the study report a positive and statistically significant relationship between the intermediation capacity of the banking system as well as its size and energy consumption. The findings also confirm a non-linear and inverted U-shaped relationship between financial development and energy demand for the MENA region. This implies that initially energy demand increases with financial development and then, at a turning point of financial development, it declines. The policy implications of these results are discussed.

The productivity cost of power outages for manufacturing small and medium enterprises in Senegal

This paper investigates on the productivity effects of power outages on manufacturing small and medium enterprises (SMEs) in Senegal, using a panel data on manufacturing firms. Productivity is estimated using stochastic frontier models, and power outages measured by their frequency or their duration. We controlled for firms owning a generator, relevant covariates as data availability permits as well. The main results are drawn from random effects linear panel model. Nonetheless, the results remain consistent to the robustness checks using different models: a double-sided truncated data model and a generalized linear model, and different productivity measures, using data envelopment analysis. We find that power outages have negative significant effects on the productivity of SMEs in Senegal. Further, firms with a generator were successful in countering the adverse effect of power outages on productivity, this make the negative effect bore only by SMEs, which in some cases cannot afford to own a generator. As a matter of fact, the manufacturing sector lost up to around 15% of the actual productivity due to power outages in 2012, and small and median firms have lost, respectively, around 4.7 and 4.2%. Besides, another important finding is the significant positive effect of access to credit on productivity. At last, it is confirmed that productivity increases with firms’ size.

Environmental Regulation Intensity, Foreign Direct Investment, and Green Technology Spillover—An Empirical Study

Many researchers have studied the relationships among heterogeneous foreign direct investment (FDI), environmental regulation, and green total factor productivity. However, no research has been done on how different types of FDI can result in green technology spillover under different levels of environmental regulation intensity. To address this research gap, in this paper, we build a static linear panel model, a static panel threshold model, and a dynamic panel threshold model to investigate the environmental regulatory threshold effect of labor-based FDI and capital-based FDI in terms of their green technology spillover. Based on the measurement of green total factor productivity (GTFP) of 36 industry sectors in China from 2003 to 2015, we first compare the threshold effects of environmental regulation on green technology spillover between labor-based FDI and capital-based FDI with a static linear model and a static threshold model. The results show that environmental regulation is unable to significantly promote the green technology spillover of labor-based FDI. However, intensifying environmental regulation can reduce the negative impact of labor-based FDI on GTFP. The effect of environmental regulation on green technology spillover of capital-based FDI is more complex. In the static linear model, environmental regulation can significantly promote the green technology spillover of capital-based FDI. In the static threshold model, the green technology spillover of capital-based FDI exists only when the environmental regulation intensity is sufficiently low or sufficiently high. Finally, the dynamic threshold model is adopted for robustness check. The results show when the environmental regulation intensity is higher than a threshold, both types of FDI can indeed result in green technology spillover. In short, our results prove that to ensure that FDI results in green technology spillover, it is necessary to continue to strengthen environmental regulation.

The effect of productivity on liquidity under financial frictions

Productivity is something that can affect the distribution of the characteristics of the assets. In this study, we investigate whether manufacturing firms in Indonesia that have high productivity have a high level of liquidity. This study uses data from manufacturing firms listed on the Indonesia Stock Exchange in the period of 2008 to 2017. We estimate the productivity level of the firm using the Generalized Method of Moments (GMM) and effects of productivity on the liquidity of the firm using the linear panel model. Results show that manufacturing firms in Indonesia with high productivity levels tend to have a higher level of liquidity than firms with lower levels of productivity. Even if Indonesia already adopt a market-based financial system, other types of financial frictions cause that firms allocate more of their resource to liquid assets than to fixed assets. Even though the effects of misallocation became weaker, misallocations of resources in manufacturing firms Indonesia are still found from a robustness test. JEL Classifications : G15, G31, G32 DOI : https://doi.org/10.26905/jkdp.v23i2.3191

Xtserialpm: A portmanteau test for serial correlation in a linear panel model

We introduce the command xtserialpm to perform the portmanteau test developed in Jochmans (2019). The procedure tests for serial correlation in the errors of a linear panel model after estimation of the regression coefficients by the within-group estimator. The test is different from the test of Inoue and Solon (2006) that is performed by xtistest (Wursten 2018) in that it allows for heteroskedasticity. In simulations documented below, xtserialpm is found to provide a much more powerful test than xtistest. xtserialpm can deal with unbalanced panel data.

The Ripple Effect of Firm-Generated Content for New Movie Releases

Marketers create social media, in the form of firm-generated content (FGC), to ignite interest in new products such as movies; in turn, there is a clear need to understand whether and how FGC influences demand. With a descriptive study, the authors investigate two potential mechanisms by which FGC may drive box office revenues: (1) a direct mechanism, such that users who see FGC directly drive revenue, and (2) a “ripple effect,” by which FGC increases movie-related user-generated content (UGC), which then drives consumption. The study data include 135,153 firm-generated and 4.9 million user-generated Twitter posts associated with 159 movies released by major studios in 2014 and 2015. Linear panel models reveal a positive, significant effect of FGC on movie sales, which UGC fully mediates, in support of a ripple effect. The model accounts for endogenous firm actions, unobserved movie heterogeneity, competition, and advertising. The evidence in support of ripple effects suggests that movie executives should not focus solely on gaining followers but rather on creating FGC that sparks conversations among users.

IMPACT OF THE CAPITAL STRUCTURE ON THE VALUE OF US AND SERBIAN FIRMS / Uticaj strukture kapitala na vrijednost američkih i srpskih firmi

According to the Modigliani–Miller theorem, the value of the company does not depend on the capital structure (debt to equity ratio). We checked the theorem in two financial markets, the global one and the regional one. In the research, we used a multiple linear regression model, panel model and VAR. The theorem did not withstand an empirical check on the financial markets of the United States and the Republic of Srpska. The value of American and Serbian firms is not indifferent to the movement in the capital structure. The growth of debt in relation to the capital in the USA and the Republic of Srpska diminishes the value of firms, i.e., value of the PE ratio. The inverse relationship between the capital structure and the value of capital is economically logical and expected, because the growth of indebtedness increases the risk of liquidity and solvency. The growth of indebtedness influences reducing of the demand for company shares and reducing of the value of the firm, i.e. reduction of the PE ratio.