Editorial: Application of Big Data, Deep Learning, Machine Learning, and Other Advanced Analytical Techniques in Environmental Economics and Policy

Editorial on the Research Topic
Application of Big Data, Deep Learning, Machine Learning, and Other Advanced Analytical Techniques in Environmental Economics and Policy

Environmental science has attracted the attention of more and more researchers around the globe, yet most of the analyses are based on traditional analytical techniques. It is noteworthy that although big data, deep learning, and other machine learning techniques have been applied in many different disciplines, including engineering, computer science, and medical science, these state-of-the-art analytical techniques have not been applied widely in the field of environmental science, nor the areas of environmental economics and management. There are only a few articles based on machine learning, for example, Magazzino et al. (2021a), Magazzino et al. (2021b), Magazzino et al. (2021c), and Magazzino et al. (2021d). Given the powerful capability of these techniques and the increasing availability of big data, their application can supplement existing research by providing a new perspective on environmental economics and management, and providing accurate forecasts and pragmatic policy suggestions.

To fill this gap, the International Society for Energy Transition Studies (ISETS) collaborated with Frontiers in an attempt to promote the application of big data, deep learning, machine learning, and other advanced analytical techniques in analyzing environmental economics and policy by inviting members of the ISETS and other non-member researchers to contribute to a dedicated research topic. The objectives are to facilitate the engagement and the advancement of research centred around energy systems. There are four participating journals for this research topic, namely, Frontiers in Energy Research, Frontiers in Environmental Science, Frontiers in Ecology and Evolution, and Frontiers in Earth Science.

The goal of this research topic is to re-examine important environmental economics and management issues by employing cutting-edge research methods based on big data, deep learning, and other machine learning techniques, as well as other advanced analytical methods. Given that many important issues in environmental economics and management are exceptionally complex, and the underlying relationships with the determinants are nonlinear, applying these Frontier research methods may prove particularly valuable because of their capability in modelling various complex and nonlinear relationships.

Thirty-six articles based on significant environmental economics and management issues were published under this research topic. All the analyses are based on state-of-the-art analytical techniques. Most of the authors addressed critical issues from an empirical and quantitative point of view by revisiting the issues with the application of big data, deep learning, other machine learning techniques, as well as other Frontier techniques. Some authors compared the findings derived from existing research studies based on traditional analytical methods with the proposed Frontier research methods. Moreover, many authors delved into burning issues or heated debates to provide insights into environmental economics and management policy formulation. The articles published under this research topic can be broadly divided into two major areas: environmental protection and energy. However, the focuses of the articles are varied and include many important issues in the two areas.

For the area of environmental protection, some researchers conducted research on the relationship between environmental protection and growth. Xiang et al. examined the impact of economic growth on carbon emission in BRICS countries by using the multivariate wavelet analysis. Khan and Wang examined the short and long-run effects of poverty, income inequality, population, and GDP per capita on carbon emission in Pakistan by applying the Autoregressive Distributive Lag (ARDL) and Non-linear Autoregressive Distributive Lag (NARDL) co-integration approach. Li et al. investigated the relation between carbon emissions and economic growth, industry structure, urbanization, research and development (R&D) investment, use of foreign capital, and growth rate of energy consumption in China based on machine learning.

Other researchers focus on emission analysis. For example, Shum et al. investigated the relative importance of carbon emissions drivers in China by employing the Least Absolute Shrinkage and Selection Operator (LASSO) model in ranking the relative importance of the independent variables. In addition, Ma et al. evaluated the feasibility of using machine learning in carbon emission analysis by employing the Gaussian Process Regression (GPR) algorithm.

Moreover, many authors conducted analyses on policy formulation. Zhang et al. examined carbon neutrality policies and technologies by adopting the scientometric analysis. Shao et al. studied the impact of environmental regulation on industrial structure upgrading by using the Pollution Information Transparency Index (PITI) to measure environmental regulation (ER) and examined the effect of ER on industrial structure upgrading. Feng et al. examined the impact of environmental regulations on China’s green total factor productivity by using econometrics analysis and machine learning. Xiao et al. investigated the effect of the green credit policy implemented by the Chinese government on firm-level industrial pollutant emissions by employing a quasi-natural experiment, propensity score matching and the difference-in-difference approach (PSM-DID). Yang et al. studied the relationship between financial inclusion and carbon reduction in Chinese counties. Wu et al. examined the impacts of the new urbanization pilot policy on air quality and related air pollutants. Wang et al. implemented a sharp regression discontinuity (RD) design and assessed air quality control effectiveness in China based on the high-volume big data acquired from 173 cities. Zheng and He evaluated the impacts of two revisions of China’s environmental protection fee on firm performance based on evidence from the stock markets.

Some researchers focus on the emissions of specific industries. For example, Li et al. examined the change in China’s construction industry’s domestic carbon emission intensity and analyzed the reason behind the change. Chen et al. studied dynamic supervision and control of volatile organic compounds (VOCs) emission from China’s furniture manufacturing industry based on big data and the internet of things (IoT).

Given that environmental issue is a significant concern to many countries, some researchers focus on the Belt and Road Initiative (BRI) countries and aim to evaluate the impacts of BRI on various environmental issues. Li et al. offered an evolutionary and counterfactual baseline to assess the environmental impact of BRI based on the distribution dynamics approach and the mobility probability plots (MPPs). In addition, Lu et al. analyzed the environmental risk contagion relations among the BRI countries and the characteristics of their network structure by using social network analysis (SNA).

Other environmental issues are also examined. For example, Xu et al. examined the environmental efficiency of grain production and its spatial effects in China’s major grain production areas by the global super-efficiency SBM model and the Spatial Durbin model.

For the field of energy, many researchers focus on electricity. Li and Cao compared the effectiveness of information feedback between emailing electricity bills to households and installing smart meters in promoting electricity conservation by employing empirical survey data from the Chinese General Social Survey and the propensity score matching method. Jin et al. analyzed the effects of sensitive information disclosure and compared the market-clearing results under different scenarios in the Chinese electricity market. He and Gao developed a dual-sector dynamic equilibrium model, and they introduced electricity consumption and water consumption in a growth model using a time series data set from 1950 to 2014 in Guangzhou, China. Jin et al. studied the way to effectively promote compliance management in the electricity market by using an evolutionary game model under two different scenarios, i.e., the scenario without governmental supervision and the scenario with governmental supervision, and explicitly described the strategic behaviours and dynamic evolution process of power enterprises and regulators in the power market. As the mismatch between energy distribution and power load in China can be alleviated by inter-regional and inter-provincial power transactions, Wang et al. studied a method to deal with inter-regional and inter-provincial transaction settlement deviation quantity based on the kernel density–entropy weight approach.

Other authors examined the efficiency, and Li et al. explored the evolution of manufacturing green development efficiency in the Yangtze River Economic Belt by considering the resource inputs and undesirable outputs in the production process using the WSR methodology, the super-SBM model, and the Tobit model. Liu et al. analyzed the impact of government corruption on energy efficiency (EE) in China from the perspective of energy regulations through statistical methods.

Energy markets has also been studied, for example, Xue et al. analyzed the dynamic trading network structure of the international crude oil and gas market by employing the dynamical similarity analysis at different time scales by inducing a multiscale embedding for dimensionality reduction. Duan et al. studied how the uncertainties and risks of the overseas oil and gas investment environment changed over time and revealed the specific occurrence probabilities of risk on different levels.

Many authors conduct research on the energy transition issue. Using the Bayesian dynamic game model, Liu et al. analyzed the tripartite coordinated regulation for the manufacturers, consumers, and governments in the new energy vehicle (NEV) market. Yu et al. employed social network analysis and patent citation information of hydrogen fuel cell vehicle-related invention patents to construct China’s hydrogen fuel cell vehicle innovation network.

As would be expected, renewable energy is also a subject in this research topic. Lu et al. studied the causal relationship between renewable energy consumption and economic growth in four countries: Brazil, Germany, Japan, and the United States, by using the recent vector autoregression (VAR)-based Granger-causality test of Rossi-Wang. In addition, Lu et al. investigated the effects of age dependency ratio and urbanization on renewable and non-renewable energy consumption in Brazil, India, China, and South Africa.

Some other energy issues are also explored in this research topic. Zhang et al. examined how target-based performance evaluation affects the accuracy of energy-saving data and found that the accuracy of the indicator improves after the central government has included energy intensity in the performance appraisal system for local officials. Xu et al. evaluated the potential of using an attentional-based LSTM network (A-LSTM) to predict heating, ventilation, and air-conditioning (HVAC) energy consumption in practical applications.

Besides the articles above, many researchers conduct research on other important topics. For example, Meseguer-Sánchez et al. analyzed the degree of advancement of the circular economy in the scientific field through a bibliometric analysis. Huang et al. studied the proximity effects of different types and sizes of urban blue spaces on property value in Changsha metropolis, China. They examined the spatial quantile effect across housing prices by using the two-stage instrumental method (2SLS) hedonic model and spatial quantile regression (SQR). Pascual and Pascual explored the cloud forests by running simulations from a predictive model which is based on artificial intelligence, satellite images, and cloud technology.

Table 1 shows a summary of the methodologies employed by different authors and it can be observed that many researchers employed big data, deep learning, and other machine learning techniques as well as other advanced analytical methods in analyzing environmental protection, energy, environmental economics and management. The application of these state-of-the-art analytical techniques offers a new perspective on many significant issues and contributes significantly to the literature.

TABLE 1

TABLE 1. Summary of the methodologies employed in the articles.

The Topic Editors, namely Tsun Se Cheong, Xunpeng Shi, Yanfei Li, and Yongping Sun, would like to express their gratitude to the authors for their contribution to this research topic and look forward to working with the research community in research collaboration soon again in the coming future.

Author Contributions

All authors listed have made a substantial, direct, and intellectual contribution to the work and approved it for publication.

Conflict of Interest

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

Publisher’s Note

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.

References

Magazzino, C., Mele, M., and Morelli, G. (2021a). The Relationship between Renewable Energy and Economic Growth in a Time of Covid-19: A Machine Learning Experiment on the Brazilian Economy. Sustainability 13 (3), 1285. doi:10.3390/su13031285

CrossRef Full Text | Google Scholar

Magazzino, C., Mele, M., and Schneider, N. (2021b). A D2C Algorithm on the Natural Gas Consumption and Economic Growth: Challenges Faced by Germany and Japan. Energy 219, 119586. doi:10.1016/j.energy.2020.119586

CrossRef Full Text | Google Scholar

Magazzino, C., Mele, M., Morelli, G., and Schneider, N. (2021c). The Nexus between Information Technology and Environmental Pollution: Application of a New Machine Learning Algorithm to OECD Countries. Util. Policy 72, 121056. doi:10.1016/j.jup.2021.101256

CrossRef Full Text | Google Scholar

Magazzino, C., Mele, M., Schneider, N., and Shahbaz, M. (2021d). Can Biomass Energy Curtail Environmental Pollution? A Quantum Model Approach to Germany. J. Environ. Manag. 287, 112293. doi:10.1016/j.jenvman.2021.112293

CrossRef Full Text | Google Scholar