Package index • lossratio

Input layer

Validation, coercion, and helpers for raw experience data.

check_experience(): Check an experience dataset
is_experience(): Check whether an object is an Experience
as_experience(): Coerce a dataset to an Experience object
add_experience_period(): Add standard period variables to an experience dataset
validate_triangle(): Validate triangle structure before building a development

Aggregation builders

Three frameworks for viewing the same long-format experience data — cohort × dev (Triangle), calendar period (Calendar), or portfolio total (Total).

build_triangle(): Build a development structure from experience data
build_calendar(): Build a calendar-based development structure from experience data
build_total(): Build a total development summary from experience data

Age-to-age (ATA) factors

Building blocks of the chain-ladder method.

build_ata(): Build age-to-age (ata) factors from Triangle data
fit_ata(): Fit age-to-age development factors
summary(<ATA>): Summarise age-to-age factor statistics
find_ata_maturity(): Find ata maturity by group

Exposure-driven (ED) intensity

Building blocks of the exposure-driven method.

build_ed(): Build exposure-driven development data
fit_ed(): Fit ED intensity factors
summary(<ED>): Summarise ED intensity statistics

Projection

Chain ladder and loss-ratio projection. fit_lr supports three methods — "sa" (stage-adaptive, default), "ed", and "cl".

fit_cl(): Fit chain ladder projection from a Triangle object
fit_lr(): Fit loss ratio projection model

Regime detection

Structural change detection across underwriting cohorts.

detect_cohort_regime() print(<CohortRegime>) summary(<CohortRegime>) print(<summary.CohortRegime>): Detect structural regime shifts across underwriting cohorts

Backtest

Hold out the latest calendar diagonals from a triangle, refit, and compare projections against the withheld actuals.

backtest() print(<Backtest>) summary(<Backtest>) print(<summary.Backtest>): Backtest a loss-ratio / chain ladder fit on existing data

Loss ratio convergence detection

Detect the development period ( $k^{**}$ ) from which the projected loss ratio stops revising and converges.

find_lr_convergence(): Find the development period at which the loss ratio estimate stabilises

Visualisation

plot() (base generic) and plot_triangle() (lossratio generic) dispatch on the object class.

plot_triangle(): Triangle plot generic
plot(<ATA>): Plot age-to-age factor diagnostics
plot(<ATAFit>): Plot an ata fit
plot(<Backtest>): Plot a backtest object
plot(<CLFit>): Plot a chain ladder fit
plot(<Calendar>): Plot calendar-based development statistics
plot(<CohortRegime>): Plot a cohort regime detection result
plot(<ED>): Plot ED intensity diagnostics
plot(<EDFit>): Plot an ED fit
plot(<LRConvergence>): Plot the LRConvergence diagnostic
plot(<LRFit>): Plot a loss ratio fit
plot(<Total>): Plot a Total object as a per-group bar chart
plot(<Triangle>): Plot development trajectories with optional summary overlay
plot_triangle(<ATA>): Plot ata factors as a triangle heatmap table
plot_triangle(<ATAFit>): Triangle heatmap for an ata fit
plot_triangle(<Backtest>): Triangle heatmap of backtest AEG
plot_triangle(<CLFit>): Plot chain ladder results as a triangle table
plot_triangle(<ED>): Plot ED intensities as a triangle heatmap table
plot_triangle(<EDFit>): Triangle heatmap for an ED fit
plot_triangle(<LRFit>): Plot loss ratio projection as a triangle heatmap
plot_triangle(<Triangle>): Plot development values as a triangle table

Other S3 methods

print / summary / longer methods registered on package classes.

backtest() print(<Backtest>) summary(<Backtest>) print(<summary.Backtest>): Backtest a loss-ratio / chain ladder fit on existing data
detect_cohort_regime() print(<CohortRegime>) summary(<CohortRegime>) print(<summary.CohortRegime>): Detect structural regime shifts across underwriting cohorts
print(<ATAFit>): Print an ATAFit object
print(<CLFit>): Print a CLFit object
print(<EDFit>): Print an EDFit object
print(<LRFit>): Print an LRFit object
summary(<ATA>): Summarise age-to-age factor statistics
summary(<ATAFit>): Summary method for ATAFit
summary(<CLFit>): Summary method for CLFit
summary(<Calendar>): Summarise calendar-development statistics (Mean, Median, Weighted)
summary(<ED>): Summarise ED intensity statistics
summary(<EDFit>): Summary method for EDFit
summary(<LRFit>): Summary method for LRFit
summary(<Total>): Summarise a Total object
summary(<Triangle>): Summarise development statistics (Mean, Median, Weighted)

Helpers

get_recent_weights(): Recent-diagonal weights for a development triangle
longer(): Reshape an object to long form (S3 generic)

Datasets

experience: Sample loss experience data