A multi-year microlevel collective risk model

For a typical insurance portfolio, the claims process for short period, typically one year, is characterized by observing frequency of together with associated severities. The collective risk model describes this portfolio as random sum aggregation claim amounts. In classical framework, simplicity, and severities are assumed to be mutually independent. However, there growing interest in relaxing independence assumption which more realistic useful practical ratemaking. While common thread has been capturing dependence between aggregate severity within single work Oh et al. (2021) provides an interesting extension addition among individual paper, we extend these works framework where have microlevel frequencies multiple years. This allows us develop factor copula that captures various types over It therefore clear earlier on one-year dependent frequency-severity models effects serial claims. We focus results using family elliptical copulas dependence. paper further how calibrate proposed illustrative data arising from Singapore company. estimated provide strong evidences all forms dependencies captured our model.