Detecting correlation between extreme probability events

G. Coletti; L.C. van der Gaag; D. Petturiti

Detecting correlation between extreme probability events

G. Coletti
, L.C. van der Gaag
, D. Petturiti

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › Academic › peer-review

Abstract

Since the classical definitions of correlation give rise to counterintuitive findings
for extreme probability events, we build upon the concept of coherent
conditional probability to introduce enhanced notions of correlation. Our new
notions allow handling extreme events in a principled way by accommodating
the different levels of strength of the zero probabilities involved. Where
the detection of correlations by means of these levels is computationally challenging, we provide a full characterisation of the correlations between extreme
probability events without reference to the complex structure of probability.

Original language	English
Title of host publication	Proceedings of the 11th Workshop on Uncertainty Processing
Subtitle of host publication	WUPES’18 Trebon, Czech Republic
Editors	V. Kratochvíl, J. Vejnarová
Place of Publication	Prague
Publisher	MatfyzPress
Pages	37-48
Number of pages	12
ISBN (Electronic)	978-80-7378-361-7
Publication status	Published - 2018

Access to Document

DetectingFinal published version, 391 KBLicence: Taverne

Cite this

@inproceedings{ef7b8d98e3a04415a1d77fa17c4cadac,

title = "Detecting correlation between extreme probability events",

abstract = "Since the classical definitions of correlation give rise to counterintuitive findings for extreme probability events, we build upon the concept of coherent conditional probability to introduce enhanced notions of correlation. Our new notions allow handling extreme events in a principled way by accommodating the different levels of strength of the zero probabilities involved. Where the detection of correlations by means of these levels is computationally challenging, we provide a full characterisation of the correlations between extreme probability events without reference to the complex structure of probability.",

author = "G. Coletti and \{van der Gaag\}, L.C. and D. Petturiti",

year = "2018",

language = "English",

pages = "37--48",

editor = "V. Kratochv{\'i}l and J. Vejnarov{\'a}",

booktitle = "Proceedings of the 11th Workshop on Uncertainty Processing",

publisher = "MatfyzPress",

}

Detecting correlation between extreme probability events. / Coletti, G.; van der Gaag, L.C.; Petturiti, D.
Proceedings of the 11th Workshop on Uncertainty Processing: WUPES’18 Trebon, Czech Republic. ed. / V. Kratochvíl; J. Vejnarová. Prague: MatfyzPress, 2018. p. 37-48.

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › Academic › peer-review

TY - GEN

T1 - Detecting correlation between extreme probability events

AU - Coletti, G.

AU - van der Gaag, L.C.

AU - Petturiti, D.

PY - 2018

Y1 - 2018

N2 - Since the classical definitions of correlation give rise to counterintuitive findings for extreme probability events, we build upon the concept of coherent conditional probability to introduce enhanced notions of correlation. Our new notions allow handling extreme events in a principled way by accommodating the different levels of strength of the zero probabilities involved. Where the detection of correlations by means of these levels is computationally challenging, we provide a full characterisation of the correlations between extreme probability events without reference to the complex structure of probability.

AB - Since the classical definitions of correlation give rise to counterintuitive findings for extreme probability events, we build upon the concept of coherent conditional probability to introduce enhanced notions of correlation. Our new notions allow handling extreme events in a principled way by accommodating the different levels of strength of the zero probabilities involved. Where the detection of correlations by means of these levels is computationally challenging, we provide a full characterisation of the correlations between extreme probability events without reference to the complex structure of probability.

M3 - Conference contribution

SP - 37

EP - 48

BT - Proceedings of the 11th Workshop on Uncertainty Processing

A2 - Kratochvíl, V.

A2 - Vejnarová, J.

PB - MatfyzPress

CY - Prague

ER -

Detecting correlation between extreme probability events

Abstract

Access to Document

Fingerprint

Cite this