Estimating the extremal index through the tail dependence concept
The extremal index θ is an important parameter in extreme value analysis when extending results from independent and identically distributed sequences to stationary ones. A connection between the extremal index and the tail dependence coefficient allows the introduction of new estimators. The propos...
| Main Author: | |
|---|---|
| Format: | article |
| Language: | eng |
| Published: |
2015
|
| Subjects: | |
| Online Access: | http://hdl.handle.net/1822/46967 |
| Country: | Portugal |
| Oai: | oai:repositorium.sdum.uminho.pt:1822/46967 |
| Summary: | The extremal index θ is an important parameter in extreme value analysis when extending results from independent and identically distributed sequences to stationary ones. A connection between the extremal index and the tail dependence coefficient allows the introduction of new estimators. The proposed ones are easy to compute and we analyze their performance through a simulation study. Comparisons with other existing methods are also presented. Case studies within environment are considered in the end. |
|---|