Circular data arise as directions, rotations, axes, clock, or calendar measurements. Applications are found in industry, envirometrics, Earth sciences and many other fields. Detecting outliers is an important problem that has been studied in several research areas. In this study, an outlier identification procedure for circular data is suggested. The proposed method is based on robust estimates of distribution parameters on the circle and it is illustrated through two real data examples.
Anomaly detection in circular data
Houyem Demni
;Giovanni C. Porzio
2023-01-01
Abstract
Circular data arise as directions, rotations, axes, clock, or calendar measurements. Applications are found in industry, envirometrics, Earth sciences and many other fields. Detecting outliers is an important problem that has been studied in several research areas. In this study, an outlier identification procedure for circular data is suggested. The proposed method is based on robust estimates of distribution parameters on the circle and it is illustrated through two real data examples.File in questo prodotto:
| File | Dimensione | Formato | |
|---|---|---|---|
|
DemniandPorzio2023_SEAS.pdf
accesso aperto
Descrizione: Contributo in volume
Licenza:
Creative commons
Dimensione
614.31 kB
Formato
Adobe PDF
|
614.31 kB | Adobe PDF | Visualizza/Apri |
I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.
