Experience rating of a layer of limit excess of
deductible: the past large losses are brought to the money of the rating
year, put through the layer as if it had been in force, developed to ultimate
and divided by the premium of their year, brought to the rating year's rates.
Usage
burning_cost(
losses,
premium,
deductible,
limit,
aggregate_deductible = 0,
aggregate_limit = Inf,
inflation = 0,
rating_year = NULL,
development = 1,
premium_index = 1,
threshold = NULL,
exclude = NULL,
weights = c("premium", "equal"),
rating_premium = NULL
)Arguments
- losses
A data frame of the large losses with the columns
yearandamount(in the money of their year); other columns, such as a claim number, are kept.The subject premium of each year: a data frame with the columns
yearandpremium, or numbers named by year. Its years are those of the experience; a year without losses has none in the layer.- deductible, limit
The layer,
limit(orInf) excess ofdeductible, for each loss.- aggregate_deductible, aggregate_limit
The annual aggregate deductible and limit of the layer's losses; the aggregate limit is often the limit times one plus the number of reinstatements.
- inflation
An annual rate (0.05 for 5%) or index values named by year, including the rating year.
- rating_year
The year being priced; by default the year after the last.
- development
One factor for all years, one per year, or factors named by year, applied to each year's layer losses. Defaults to 1.
The factor that brings each year's premium to the rating year's rates and exposure, as
development. Defaults to 1.- threshold
The reporting threshold of the losses, used for the warning; NULL (the default) for none.
- exclude
Years left out of the burning cost (they are still shown).
- weights
"premium" (the default) for the total layer losses over the total on-level premium, or "equal" for the mean of the years' burning costs.
The subject premium of the rating year; the expected layer loss is the burning cost times it. NULL (the default) gives no expected loss.
Value
An object of class netsimr_burning_cost, a list with
by_year (each year's number of losses and of losses in the layer, its
indexed losses, layer losses, development, developed and aggregate layer losses,
premium, premium index, on-level premium, burning cost and whether it is
included, and, with rating_premium, the layer losses at the rating year's
premium), losses (each loss indexed and in the layer), rate (the
burning cost), expected_loss, rating_year, rating_premium,
n_layer_losses (in the years included) and the layer.
Details
Each loss is indexed to the rating year with inflation, and its
part in the layer, \(\min(\max(x - d, 0), l)\), taken. Each year's layer
losses are multiplied by its development factor (for the late reported
claims and the development of the known ones, e.g. from
chain_ladder), and then the aggregate deductible and limit apply.
The burning cost of a year is its layer losses over its subject premium times
its premium_index (the on-level premium). The burning cost of the layer
is the total over the total, weighting years by premium, or the mean of the
years' rates.
Losses are usually given above a reporting threshold. A loss just below the threshold, indexed, reaches the layer when the indexed threshold exceeds the deductible; a warning names the years where it does.
See also
pareto_rating, property_exposure_rating and
technical_premium.
Other reinsurance pricing functions:
credibility_weight(),
mbbefd_curve(),
pareto_rating(),
property_exposure_rating(),
technical_premium()
Examples
bc <- burning_cost(xl_large_losses, xl_premium[, c("year", "premium")],
deductible = 1e6, limit = 4e6, inflation = 0.04,
development = setNames(xl_premium$development, xl_premium$year),
premium_index = setNames(xl_premium$rate_index, xl_premium$year),
threshold = 5e5, rating_premium = 60e6)
bc
#> Burning cost of 4,000,000 xs 1,000,000 for 2025
#> Years: 10 (2015 to 2024)
#> Losses in the layer: 25
#> Burning cost: 4.972% of subject premium (premium weighted)
#> Expected layer loss: 2,983,162
bc$by_year[, c("year", "layer_losses", "onlevel_premium", "burning_cost")]
#> year layer_losses onlevel_premium burning_cost
#> 1 2015 4213705.33 38728000 0.108802554
#> 2 2016 550147.77 40800000 0.013484014
#> 3 2017 0.00 42669000 0.000000000
#> 4 2018 441448.23 45696000 0.009757150
#> 5 2019 6278006.84 48300000 0.132579026
#> 6 2020 44983.18 50774000 0.000921387
#> 7 2021 4068163.53 52624000 0.083490738
#> 8 2022 4893266.79 54796000 0.102694664
#> 9 2023 1564784.64 56358000 0.036094610
#> 10 2024 366560.00 58400000 0.010356575
