Arbeitspapier
K-expectiles clustering
K-means clustering is one of the most widely-used partitioning algorithm in cluster analysis due to its simplicity and computational efficiency, but it may not provide ideal clustering results when applying to data with non-spherically shaped clusters. By considering the asymmetrically weighted distance, We propose the K-expectile clustering and search the clusters via a greedy algorithm that minimizes the within cluster τ -variance. We provide algorithms based on two schemes: the fixed τ clustering, and the adaptive τ clustering. Validated by simulation results, our method has enhanced performance on data with asymmetric shaped clusters or clusters with a complicated structure. Applications of our method show that the fixed τ clustering can bring some flexibility on segmentation with a decent accuracy, while the adaptive τ clustering may yield better performance.
- Language
-
Englisch
- Bibliographic citation
-
Series: IRTG 1792 Discussion Paper ; No. 2021-003
- Classification
-
Wirtschaft
Mathematical and Quantitative Methods: General
- Subject
-
clustering
expectiles
asymmetric quadratic loss
image segmentation
- Event
-
Geistige Schöpfung
- (who)
-
Wang, Bingling
Li, Yingxing
Härdle, Wolfgang
- Event
-
Veröffentlichung
- (who)
-
Humboldt-Universität zu Berlin, International Research Training Group 1792 "High Dimensional Nonstationary Time Series"
- (where)
-
Berlin
- (when)
-
2021
- Handle
- Last update
-
10.03.2025, 11:43 AM CET
Data provider
ZBW - Deutsche Zentralbibliothek für Wirtschaftswissenschaften - Leibniz-Informationszentrum Wirtschaft. If you have any questions about the object, please contact the data provider.
Object type
- Arbeitspapier
Associated
- Wang, Bingling
- Li, Yingxing
- Härdle, Wolfgang
- Humboldt-Universität zu Berlin, International Research Training Group 1792 "High Dimensional Nonstationary Time Series"
Time of origin
- 2021
Other Objects (12)
On adaptive estimation in partial linear models
Semi-parametric estimation of generalized partially linear single-index models
Exploring credit data
Networks of news and cross-sectional returns
Rodeo or ascot: Which hat to wear at the crypto race?
Robust adaptive estimation of dimension reduction space
Backtesting beyond VaR
E-learning / e-teaching of statistics: Students' and teachers' views
Dynamic nonparametric state price density estimation using constrained least squares and the bootstrap
Large sample theory of the estimation of the error distribution for a semiparametric model
Flexible time series analysis
Discrete time option pricing with flexible volatility estimation
On adaptive estimation in partial linear models
Semi-parametric estimation of generalized partially linear single-index models
Exploring credit data
Networks of news and cross-sectional returns
Rodeo or ascot: Which hat to wear at the crypto race?
Robust adaptive estimation of dimension reduction space
Backtesting beyond VaR
E-learning / e-teaching of statistics: Students' and teachers' views
Dynamic nonparametric state price density estimation using constrained least squares and the bootstrap
Large sample theory of the estimation of the error distribution for a semiparametric model
Flexible time series analysis
Discrete time option pricing with flexible volatility estimation
On adaptive estimation in partial linear models
Semi-parametric estimation of generalized partially linear single-index models
Exploring credit data
Networks of news and cross-sectional returns
Rodeo or ascot: Which hat to wear at the crypto race?
Robust adaptive estimation of dimension reduction space
Backtesting beyond VaR
E-learning / e-teaching of statistics: Students' and teachers' views
Dynamic nonparametric state price density estimation using constrained least squares and the bootstrap
Large sample theory of the estimation of the error distribution for a semiparametric model
Flexible time series analysis