User: Guest  Login
Title:

Selection of Sparse Vine Copulas in High Dimensions with the Lasso

Document type:
Zeitungsartikel
Author(s):
Müller, D. and Czado C.
Abstract:
We propose a novel structure selection method for high dimensional (d > 100) sparse vine copulas. Current sequential greedy approaches for structure selection require calculating spanning trees in hundreds of dimensions and fitting the pair copulas and their parameters iteratively throughout the structure selection process. Our method uses a connection between the vine and structural equation models (SEMs). The later can be estimated very fast using the Lasso, also in very high dimensions, to ob...     »
Keywords:
Dependence Modeling, Vine Copula, Lasso, Sparsity
Dewey Decimal Classification:
510 Mathematik
Journal title:
Statistics and Computing
Year:
2019
Journal volume:
29
Year / month:
2019-03
Quarter:
1. Quartal
Month:
Mar
Journal issue:
2
Pages contribution:
269-287
Reviewed:
ja
Language:
en
Fulltext / DOI:
doi:10.1007/s11222-018-9807-5
Publisher:
Springer US
Publisher address:
New York, NY
Print-ISSN:
0960-3174
E-ISSN:
1573-1375
Status:
Erstveröffentlichung
Date of publication:
15.03.2019
TUM Institution:
Professur für Angewandte Mathematische Statistik
Format:
Text
 BibTeX