Entropy as a measure of dependency for categorized data

Skotarczak, Ewa; Dobek, A.; Moliński, Krzysztof

doi:10.2478/bile-2018-0014

Cited by 5 publications

(6 citation statements)

References 17 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Characteristics of Joe's estimator and the regression-based estimator for simulated data sets were investigated in our earlier paper (Skotarczak et al, 2018). Under many generated data scenarios it was proven that the regression approach yields more accurate results than Joe's proposal, especially for highly correlated data.…”

Section: Discussionmentioning

confidence: 99%

“…The values of b1 and b0 for two-row tables with two, three, four and five columns are given in a paper by Skotarczak et al (2018).…”

Section: Methodsmentioning

confidence: 99%

“…Pearson's correlation and a relative entropy measure of multivariate dependence. Skotarczak et al (2018) developed a regression-based approach to the assessment of correlation coefficients using normalized mutual information. Simulated studies conducted for two-row contingency tables demonstrated some advantages of the regression approach over Joe's estimator.…”

Section: Described the Relation Betweenmentioning

confidence: 99%

See 2 more Smart Citations

Comparison of some correlation measures for continuous and categorical data

Skotarczak

Dobek

Moliński

2019

Biometrical Letters

Self Cite

View full text Add to dashboard Cite

SummaryIn the literature there can be found a wide collection of correlation and association coefficients used for different structures of data. Generally, some of the correlation coefficients are conventionally used for continuous data and others for categorical or ordinal observations. The aim of this paper is to verify the performance of various approaches to correlation coefficient estimation for several types of observations. Both simulated and real data were analysed. For continuous variables, Pearson’s r2 and MIC were determined, whereas for categorized data three approaches were compared: Cramér’s V, Joe’s estimator, and the regression-based estimator. Two method of discretization for continuous data were used. The following conclusions were drawn: the regression-based approach yielded the best results for data with the highest assumed r2 coefficient, whereas Joe’s estimator was the better approximation of true correlation when the assumed r2 was small; and the MIC estimator detected the maximal level of dependency for data having a quadratic relation. Moreover, the discretization method applied to data with a non-linear dependency can cause loss of dependency information. The calculations were supported by the R packages arules and minerva.

show abstract

Section: Discussionmentioning

confidence: 99%

“…The values of b1 and b0 for two-row tables with two, three, four and five columns are given in a paper by Skotarczak et al (2018).…”

Section: Methodsmentioning

confidence: 99%

Section: Described the Relation Betweenmentioning

confidence: 99%

See 1 more Smart Citation

Comparison of some correlation measures for continuous and categorical data

Skotarczak

Dobek

Moliński

2019

Biometrical Letters

Self Cite

View full text Add to dashboard Cite

show abstract

“…According to the characteristics of entropy, we can determine the randomness and disorder degree of an event by calculating the entropy value, or the entropy value can be applied to judge the dispersion degree of an indicator. The greater the dispersion degree of an indicator, the greater the influence of this indicator on the comprehensive evaluation (Skotarczak et al, 2018). Therefore, the weight of each indicator can be calculated according to the variation degree of each indicator, using information entropy as a tool to provide the basis for a comprehensive evaluation of multiple indicators (Zhou and Yang, 2019).…”

Section: Entropy Methodsmentioning

confidence: 99%

Temporal variability of social vulnerability to storm surges in Shenzhen, China

Shen

Kelly³

et al. 2019

Preprint

View full text Add to dashboard Cite

Abstract. A social vulnerability evaluation of storm surges is important for any coastal city to commence in order to provide marine disaster preparedness and mitigation procedures and to formulate post-disaster emergency plans for coastal communities. This study establishes an integrated evaluation system of social vulnerability by blending a variety of single evaluation methods, applying the idea of combination weighting and calculating the social vulnerability index of storm surges. Shenzhen, with a current reputation of having the most economic development potential and a representative city in China, is chosen to evaluate its social vulnerability to storm surges via a historical social and economic statistical dataset spanning from 1986 to 2016. The research extends further by analyzing the city’s temporal variability. Results reveal that social vulnerability keeps almost constant from 1986–1991 and 1993–2004, while it decreased sharply in the remainder of times to show a ‘stair-type’ declining curve over the past 30 years. Resilience is progressively increasing by virtue of a continuous increase in medical institutions, fixed asset investments and salary levels of employees. These determinants contribute to the overall downward trend of social vulnerability for Shenzhen. Exposure and sensitivity increased slowly with some fluctuation, causing the changes of social responsibility to transpire.

show abstract

“…The maps apply the Lambert conformal conic (LCC) projection due to the country's middle-latitude presence and predominantly east-west expanse. The LCC projection offers flexibility in adjustable standard parallels for plotting at different scales, where conformality is held true, angular distortion at any parallel (except for the poles) is essentially zero, and meridians are right angles (Snyder, 1987). The LCC projection emphasizes the conceptual quality of secancy for conics and has been the conformal projection of choice for midlatitudes (Pearson II, 1990).…”

Section: Introductionmentioning

confidence: 99%

Trends in social vulnerability to storm surges in Shenzhen, China

Shen

Kelly³

et al. 2020

Nat. Hazards Earth Syst. Sci.

View full text Add to dashboard Cite

Abstract. An evaluation of social vulnerability to storm surges is important for any coastal city to provide marine disaster preparedness and mitigation procedures and to formulate post-disaster emergency plans for coastal communities. This study establishes an integrated evaluation system of social vulnerability by blending a variety of single-evaluation methods, which are subsequently combined by weighting in order to calculate a common social vulnerability index. Shenzhen has a current reputation of having considerable economic development potential and is a representative city in China. It is chosen for an evaluation of its social vulnerability to storm surges via a historical social and economic statistical dataset spanning the period 1986–2016. Exposure and sensitivity increased slowly with some fluctuation, leading to some alterations of the social vulnerability trend. Social vulnerability stayed almost constant during 1986–1991 and 1993–2004, while it decreased sharply afterwards to form a “stair-type” declining curve over the past 31 years. Resilience is progressively increasing by virtue of a continuous increase in medical services supply, fixed asset investments, and salary levels of employees. These determinants contribute to the overall downward trend of social vulnerability for Shenzhen.

show abstract

Entropy as a measure of dependency for categorized data

Cited by 5 publications

References 17 publications

Comparison of some correlation measures for continuous and categorical data

Comparison of some correlation measures for continuous and categorical data

Temporal variability of social vulnerability to storm surges in Shenzhen, China

Trends in social vulnerability to storm surges in Shenzhen, China

Contact Info

Product

Resources

About