
The indicated rate change: the loss ratio method
Source:R/pricing_monitoring.R
indicated_rate_change.RdProjects each year's ultimate losses to the period the new rates will apply to, with a claims trend, divides them by the year's on-level premium, and compares the projected loss ratio with the permissible one, which leaves room for the expenses and the profit.
Usage
indicated_rate_change(
losses,
premium,
years,
trend = 0,
effective,
rate_period = 1,
term = 1,
expense_ratio = 0,
profit = 0,
weights = NULL,
credibility = 1,
complement = NULL
)Arguments
- losses
The ultimate losses of each year, such as from
chain_ladder.The on-level earned premium of each year (
on_level_premium).- years
The years.
- trend
The annual claims trend (inflation of the loss cost), as a share.
- effective
The date the new rates take effect: a Date or a decimal year.
- rate_period
How long the new rates will be in force, in years.
- term
The length of the policies in years.
- expense_ratio
The expenses (commissions and other costs varying with the premium) as a share of the premium.
- profit
The profit and contingencies provision, as a share of the premium.
- weights
Weights of the years, such as to give the latest more weight; by default their on-level premium.
- credibility
The credibility of the experience, from 0 to 1.
- complement
The loss ratio blended with the experience when its credibility is below 1.
Value
A list with by_year (each year's losses, trend factor, trended losses,
premium and projected loss ratio), the experience loss ratio, the blended loss
ratio, the permissible loss ratio and the indicated rate change.
Details
Each year's losses are trended from the middle of the year, \(y + 0.5\)
for calendar or accident year \(y\), to the average date of the claims of the
policies the new rates will price: the date they take effect, plus half the period
they will be in force, plus half a policy term. The experience loss ratio weighs
the years by their on-level premium (or weights); with credibility
below 1 it is blended with the complement, such as the loss ratio the
current rates assume. The indicated change is that loss ratio over the
permissible loss ratio, \(1 - \textrm{expense ratio} - \textrm{profit}\), less 1.
Examples
changes <- data.frame(date = as.Date(c("2021-07-01", "2023-01-01", "2024-04-01")),
change = c(0.05, -0.02, 0.08))
onl <- on_level_premium(c(9.2e6, 9.8e6, 10.4e6, 11.1e6), 2021:2024, changes)
rate <- indicated_rate_change(c(6.3e6, 6.6e6, 7.4e6, 7.6e6), onl$on_level_premium, 2021:2024,
trend = 0.04, effective = as.Date("2026-01-01"),
expense_ratio = 0.27, profit = 0.05)
rate[c("experience_loss_ratio", "indicated_change")]
#> $experience_loss_ratio
#> [1] 0.749686
#>
#> $indicated_change
#> [1] 0.1024795
#>