
Exposure rating of property and casualty excess of loss layers
Source:R/reinsurance_pricing.R
property_exposure_rating.RdRates a layer of limit excess of deductible from a risk
profile, without loss experience. Each band's (or policy's) expected loss, its
premium times its loss ratio, is shared between the layer and the rest by an
exposure curve. property_exposure_rating() uses an MBBEFD curve of the
damage as a share of the sum insured (mbbefd_curve): the layer takes
\(G((d + l) / M) - G(d / M)\) of a band with sum insured \(M\).
casualty_exposure_rating() uses the limited expected value of a claim
size distribution: the layer takes
\((LEV(\min(P, d + l)) - LEV(\min(P, d))) / LEV(P)\) of a band with policy
limit \(P\).
Usage
property_exposure_rating(
profile,
deductible,
limit,
curve = "Y3",
loss_ratio = NULL
)
casualty_exposure_rating(
profile,
deductible,
limit,
severity,
severity_params,
loss_ratio = NULL
)Arguments
- profile
A data frame with a row per band of the risk profile (or per policy): for property, the columns
sum_insured(the band's sum insured or maximum possible loss per risk, often its average) andpremium; for casualty,policy_limitandpremium. An optionalloss_ratiocolumn gives each band's expected loss ratio; other columns, such as a band label, are kept.- deductible
The deductible of the layer, at least 0.
- limit
The limit of the layer, above 0;
Inffor an unlimited layer.- curve
The exposure curve: a Swiss Re curve by name ("Y1", "Y2", "Y3", "Y4" or "Lloyds") or by its parameter c, or an MBBEFD curve as
c(b = , g = ).- loss_ratio
The expected ground-up loss ratio: one for all bands, or one per band. NULL (the default) takes the profile's
loss_ratiocolumn.- severity
The claim size distribution of the casualty business, as in
simulate_claims: "LogNormal" (meanlog, sdlog), "Gamma" (shape, scale), "Exponential" (rate) or "Pareto" (alpha, x_m).- severity_params
Its parameters, in that order or named.
Value
An object of class netsimr_exposure_rating, a list with
bands (the profile with each band's loss ratio, expected loss, the shares
of the curve at the deductible and the exhaustion point, the layer's share and
the expected layer loss), expected_loss (the layer's), rate (the
expected layer loss as a share of the subject premium), subject_premium,
ground_up_loss, the layer and the curve or severity.
See also
burning_cost and pareto_rating, the experience
rating of a layer, and technical_premium.
Other reinsurance pricing functions:
burning_cost(),
credibility_weight(),
mbbefd_curve(),
pareto_rating(),
technical_premium()
Examples
rating <- property_exposure_rating(xl_property_profile, deductible = 2e6, limit = 8e6,
curve = "Y3", loss_ratio = 0.55)
rating
#> Exposure rating (property) of 8,000,000 xs 2,000,000
#> Curve: Swiss Re Y3 (c = 3)
#> Bands: 9
#> Subject premium: 57,227,000
#> Ground-up loss: 31,474,850
#> Expected layer loss: 2,370,411
#> Rate: 4.142% of subject premium
rating$bands[, c("band", "sum_insured", "layer_share", "layer_loss")]
#> band sum_insured layer_share layer_loss
#> 1 0 to 250k 138000 0.0000000 0.0
#> 2 250k to 500k 382000 0.0000000 0.0
#> 3 500k to 1m 750000 0.0000000 0.0
#> 4 1m to 2.5m 1715000 0.0000000 0.0
#> 5 2.5m to 5m 3638000 0.1960620 586509.6
#> 6 5m to 10m 7200000 0.3749192 748320.0
#> 7 10m to 25m 16450000 0.3896002 620340.9
#> 8 25m to 50m 34875000 0.3274763 301507.5
#> 9 50m to 100m 69000000 0.2775672 113733.2
casualty_exposure_rating(xl_casualty_profile, deductible = 1e6, limit = 4e6,
severity = "LogNormal", severity_params = c(10, 2), loss_ratio = 0.65)
#> Exposure rating (casualty) of 4,000,000 xs 1,000,000
#> Curve: Log-Normal (meanlog = 10, sdlog = 2)
#> Bands: 6
#> Subject premium: 49,000,000
#> Ground-up loss: 31,850,000
#> Expected layer loss: 3,298,205
#> Rate: 6.731% of subject premium