A Study of the Transfer Entropy Networks on Industrial Electricity Consumption

Yao, Can-Zhong; Kuang, Peng-Cheng; Lin, Qing-Wen; Sun, Bo

doi:10.3390/e19040159

Cited by 12 publications

(7 citation statements)

References 56 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Different MSAs have different patterns indicating that there are different hierarchical structures of information flow. Yao et al studied information transfer routes between crossindustry and cross-region electricity consumption data based on transfer entropy and MST and found that the MSTs follow a chain-like structure in developed provinces and starlike structures in developing provinces [69]. Following their study, we investigate evolving shapes of the yearly MSAs.…”

Section: B Yearly Evolution Of the Maximum Spanning Arborescencesmentioning

confidence: 99%

Information Flow Networks of Chinese Stock Market Sectors

et al. 2020

View full text Add to dashboard Cite

Transfer entropy measures the strength and direction of information flow between different time series. We study the information flow networks of the Chinese stock market and identify important sectors and information flow paths. This paper uses the daily closing price data of the 28 level-1 sectors from Shenyin & Wanguo Securities ranging from 2000 to 2017 to study the information transmission between different sectors. We construct information flow networks with the sectors as the nodes and the transfer entropy between them as the corresponding edges. Then we adopt the maximum spanning arborescence (MSA) to extract important information flows and the hierarchical structure of the networks. We find that, during the whole sample period, the composite sector is an information source of the whole stock market, while the nonbank financial sector is the information sink. We also find that the non-bank finance, bank, computer, media, real estate, medical biology and non-ferrous metals sectors appear as high-degree root nodes in the outgoing and incoming information flow MSAs. Especially, the non-bank finance and bank sectors have significantly high degrees after 2008 in the outgoing information flow networks. We uncover how stock market turmoils affect the structure of the MSAs. Finally, we reveal the specificity of information source and sink sectors and make a conclusion that the root node sector acts as the information sink of the incoming information flow networks. Overall, our analyses show that the structure of information flow networks changes with time and the market exhibits a sector rotation phenomenon. Our work has important implications for market participants and policy makers in managing market risks and controlling the contagion of risks.

show abstract

Section: B Yearly Evolution Of the Maximum Spanning Arborescencesmentioning

confidence: 99%

Information Flow Networks of Chinese Stock Market Sectors

et al. 2020

View full text Add to dashboard Cite

show abstract

“…Therefore, correlationbased algorithm [2] and complex network [3] have been developed as efficient data-mining tools to identify the pattern of data and to reveal industry structure information [4]. According to literature investigation [5][6] [7], the basic strategy of such study work lies in that, first, a distance or strength, undirected [8] or directed [9], is defined to quantitatively express the correlation between time series pairs and then, a topology with a certain principle of optimization is generated based on the distance/strength list to manifest the intrinsic associative structure from a macroscopic perspective. Undirected distance/strength is usually defined from Pearson correlation coefficient.…”

Section: Introductionmentioning

confidence: 99%

Interprovincial Transfer Entropy Analysis on Electricity Consumption by Secondary and Tertiary Industries

Tuo

et al. 2024

IEEE Access

View full text Add to dashboard Cite

As electricity consumption is direct feedback of the production and operation state of an industry, there should be links between lead-lag relation in electricity consumption volumes and driving/transferring effects cross regions and industries. In this paper, lead-lag relation between provincial secondary industry (SI) and tertiary industry (TI) electricity consumption volumes is studied based on transfer entropy (TE) theory. Significant TE flow from SI to TI is observed both intra-provincially and interprovincially in a large part of China and periodic fluctuations in electricity consumption curves of SI emerge stably one month ahead of that of TI. It is also found that the curves of SI electricity consumption show more significant correlation with economic cycle in GDP than that of TI. Meanwhile, in both SI and TI cases, Shanghai (SH) manifests noticeable driving effect on other provinces, especially those in Yangtze River Economic Belt. Maximum spanning arborescence topology is built to exhibit direct affecting routes with maximized TE flows of SI and TI from SH to other provinces in the Belt. It is deduced that in SI case it takes stages for the influence from SH to spread while in TI case the influence tends to directly radiate elsewhere. The study attempts to propose an approach to looking through the economy from a perspective of electricity consumption and provide a reference for making industrial strategy.

show abstract

“…Additionally, Transfer Entropy (TE) exists, which allows identification of a cause-effect relationship by not accounting for straightforward and uniquely shared information [15]. TE has been applied to many complex problems from diverse research fields e.g., oscillation analysis [16], finance [17,18], sensors [19][20][21][22], biosignals [23,24], thermonuclear fusion [25], complex networks [26], geophysical phenomena [27,28], industrial energy consumption network [29] and algorithmic theory of information [30]. In addition, TE has been implemented in non-Gaussian distributions, such as: multivariate exponential, logistic, Pareto (type I-IV) and Burr distributions [31].…”

Section: Introductionmentioning

confidence: 99%

Parallelism Strategies for Big Data Delayed Transfer Entropy Evaluation

2019

View full text Add to dashboard Cite

Generated and collected data have been rising with the popularization of technologies such as Internet of Things, social media, and smartphone, leading big data term creation. One class of big data hidden information is causality. Among the tools to infer causal relationships, there is Delay Transfer Entropy (DTE); however, it has a high demanding processing power. Many approaches were proposed to overcome DTE performance issues such as GPU and FPGA implementations. Our study compared different parallel strategies to calculate DTE from big data series using a heterogeneous Beowulf cluster. Task Parallelism was significantly faster in comparison to Data Parallelism. With big data trend in sight, these results may enable bigger datasets analysis or better statistical evidence.

show abstract

A Study of the Transfer Entropy Networks on Industrial Electricity Consumption

Cited by 12 publications

References 56 publications

Information Flow Networks of Chinese Stock Market Sectors

Information Flow Networks of Chinese Stock Market Sectors

Interprovincial Transfer Entropy Analysis on Electricity Consumption by Secondary and Tertiary Industries

Parallelism Strategies for Big Data Delayed Transfer Entropy Evaluation

Contact Info

Product

Resources

About