Risk models based on time series for count random variables
In this paper, we generalize the classical discrete time risk model by introducing a dependence relationship in time between the claim frequencies. The models used are the Poisson autoregressive model and the Poisson moving average model. In particular, the aggregate claim amount and related quantities such as the stop-loss premium, value at risk and tail value at risk are discussed within this framework.
Year of publication: |
2011
|
---|---|
Authors: | Cossette, Hélène ; Marceau, Étienne ; Toureille, Florent |
Published in: |
Insurance: Mathematics and Economics. - Elsevier, ISSN 0167-6687. - Vol. 48.2011, 1, p. 19-28
|
Publisher: |
Elsevier |
Keywords: | Discrete time risk model Dependence Poisson MA(1) process Poisson MA(q) process Poisson AR(1) process |
Saved in:
Saved in favorites
Similar items by person
-
Risk models based on time series for count random variables
Cossette, Hélène, (2011)
-
Risk models based on time series for count random variables
Cossette, Hélène, (2011)
-
Compound binomial risk model in a markovian environment
Cossette, Hélène, (2004)
- More ...