2013
DOI: 10.1016/j.ins.2012.11.026
|View full text |Cite
|
Sign up to set email alerts
|

Testing noisy numerical data for monotonic association

Abstract: Rank correlation measures are intended to measure to which extent there is a monotonic association between two observables. While they are mainly designed for ordinal data, they are not ideally suited for noisy numerical data. In order to better account for noisy data, a family of rank correlation measures has previously been introduced that replaces classical ordering relations by fuzzy relations with smooth transitions -thereby ensuring that the correlation measure is continuous with respect to the data. The… Show more

Help me understand this report

Search citation statements

Order By: Relevance

Paper Sections

Select...
1
1
1
1

Citation Types

0
13
0

Year Published

2013
2013
2019
2019

Publication Types

Select...
3
3

Relationship

1
5

Authors

Journals

citations
Cited by 20 publications
(13 citation statements)
references
References 35 publications
(46 reference statements)
0
13
0
Order By: Relevance
“…Another option, which as far as we know has not been used in the area of monitoring GWBs, is RoCoCo (robust rank correlation coefficients), proposed by Bodenhofer [76,77], who presents a set of measures to test for monotonic associations between two observables. These robust rank correlation measures are based on fuzzy orderings, and the tests developed seem to outperform the classical variants because they are more robust for small samples when facing noise.…”
Section: Nonparametric Methodsmentioning
confidence: 99%
“…Another option, which as far as we know has not been used in the area of monitoring GWBs, is RoCoCo (robust rank correlation coefficients), proposed by Bodenhofer [76,77], who presents a set of measures to test for monotonic associations between two observables. These robust rank correlation measures are based on fuzzy orderings, and the tests developed seem to outperform the classical variants because they are more robust for small samples when facing noise.…”
Section: Nonparametric Methodsmentioning
confidence: 99%
“…The presence of Granger causality [44] does not assure a cause-effect relation. It may be that a third element exists that is not represented in the data that causes this.…”
Section: Granger Causalitymentioning
confidence: 99%
“…From the point of view of the model validation, this assures there exists a relation between both time series. Going further, and considering that the model is based on time series and the model and the system must be related, a Robust Rank Correlation Coefficient and a Corresponding Test (implemented on the ROCOCO R package) can be performed [44]. This test indicates that between both time series there clearly exists a correlation, showing that the model is representing the behavior of the system at some level of correctness.…”
mentioning
confidence: 99%
“…Among the different alternatives to measure correlations, it is possible to highlight the not assume a specific parametric model or specific distributions for the data [42]. Thus, the Spearman's ρ and Kendall's τ are considered to be suited to evaluate the correlation between the surface roughness and the feed rate, cutting speed and slot width.…”
Section: Surface Roughnessmentioning
confidence: 99%