The premium of a layer from its expected loss: the total premium covers the expected loss and a loading for expenses, brokerage and profit, and with paid reinstatements part of it comes as reinstatement premium, so the upfront premium is lower.
Usage
technical_premium(
expected_loss,
limit,
loading = 0,
reinstatements = 0,
reinstatement_rate = 1,
used = NULL
)Arguments
- expected_loss
The expected loss of the layer.
- limit
The limit of the layer.
- loading
The share of the total premium for expenses, brokerage and profit, from 0 to below 1. Defaults to 0.
- reinstatements
The number of paid reinstatements; 0 (the default) for none,
Inffor unlimited.- reinstatement_rate
The rate of each reinstatement, as a share of the upfront premium. Defaults to 1 (100%).
- used
The expected reinstated cover as a share of the limit; NULL (the default) for \(\min(EL / l, n)\).
Value
A list with expected_loss, total_premium,
upfront_premium, reinstatement_premium (expected),
rate_on_line (the upfront premium over the limit), payback (the
limit over the upfront premium, in years) and used.
Details
The total premium is \(EL / (1 - loading)\). With \(n\)
reinstatements paid at the rate \(r\) of the upfront premium \(P\), pro rata
as to amount, the expected reinstatement premium is \(P r u\), where \(u\) is
the expected reinstated cover as a share of the limit,
\(E[\min(S, n l)] / l\) for the annual layer loss \(S\); so
\(P = EL / ((1 - loading)(1 + r u))\). Without used, \(u\) is taken as
\(\min(EL / l, n)\), which is exact when the annual losses never exceed the
reinstated cover; the burning cost's layer losses at the rating premium give a
better one (see the examples). For the reinstatement premiums of each simulated
year, see reinstatement_premium.
See also
Other reinsurance pricing functions:
burning_cost(),
credibility_weight(),
mbbefd_curve(),
pareto_rating(),
property_exposure_rating()
Examples
technical_premium(expected_loss = 1.2e6, limit = 4e6, loading = 0.25,
reinstatements = 1, reinstatement_rate = 1)
#> $expected_loss
#> [1] 1200000
#>
#> $total_premium
#> [1] 1600000
#>
#> $upfront_premium
#> [1] 1230769
#>
#> $reinstatement_premium
#> [1] 369230.8
#>
#> $rate_on_line
#> [1] 0.3076923
#>
#> $payback
#> [1] 3.25
#>
#> $used
#> [1] 0.3
#>
# the reinstated cover used in each past year, at the rating year's premium
bc <- burning_cost(xl_large_losses, xl_premium[, c("year", "premium")], 1e6, 4e6,
inflation = 0.04, rating_premium = 60e6)
used <- mean(pmin(bc$by_year$at_rating_premium, 1 * 4e6)) / 4e6
technical_premium(bc$expected_loss, 4e6, loading = 0.25, reinstatements = 1, used = used)
#> $expected_loss
#> [1] 2782345
#>
#> $total_premium
#> [1] 3709794
#>
#> $upfront_premium
#> [1] 2494674
#>
#> $reinstatement_premium
#> [1] 1215120
#>
#> $rate_on_line
#> [1] 0.6236686
#>
#> $payback
#> [1] 1.603416
#>
#> $used
#> [1] 0.4870854
#>
