Skip to main content

methods.OneYearCDRResult

What one_year_cdr() returns: pyarrow Tables with fixed column types.

Usage

Source

methods.OneYearCDRResult(
    method,
    as_of,
    dev_grain_months,
    n_draws,
    seed,
    origins,
    totals,
    quantiles,
    draws
)

A missing number is an Arrow null, never NaN. to_polars(name) gives any of the tables as a polars DataFrame.

Every simulated number is about ultimate_change, next year’s re-estimated ultimate minus today’s: positive when the reserve is strengthened, negative when it is released. That is the opposite sign of the claims development result in kernels.simulate_one_year_cdr, which is positive for a release.

Attributes

method: str

"one_year_cdr".

as_of: datetime.date

The information date, the evaluation date of the latest cell.

dev_grain_months: int

Months per development step, always 12.

n_draws: int

The number of simulated years.

seed: int or None

The seed as passed; None means the draws came from fresh entropy and cannot be made again.

origins: pyarrow.Table

One row per origin period: origin and origin_period (as in ReserveResult), latest_dev_lag (int64), latest, ultimate and ibnr (today’s chain-ladder figures), mean_ultimate_change and sd_ultimate_change (the draws’ mean, and their standard deviation with ddof=1, null when n_draws is 1), cdr_se (the Merz-Wuthrich standard error of the one-year claims development result) and runoff_se (Mack’s standard error of the whole run-off, what mack() reports as mack_se under the same sigma_rule and zero_cells), all float64. An origin at its last development age has a change of 0 in every draw.

totals: pyarrow.Table

One row: latest, ultimate, ibnr, mean_ultimate_change, sd_ultimate_change, cdr_se, runoff_se (float64) and n_draws (int64), for the sum over the origins. The standard errors are not the sums of the origins’: the origins share the factors.

quantiles: pyarrow.Table

One row per origin, or the total, and level: origin and origin_period (null for the total), level (the probability, as passed), ultimate_change (the quantile of the draws, numpy’s linear rule, which is np.percentile with the level times 100) and tvar (the mean of the draws at or above that quantile). The total’s rows come first, then each origin’s in origin order, each in the order of the levels.

draws: pyarrow.Table
Every draw of every origin, n_draws times the number of origins rows, draw by draw: draw (int64, from 0), origin, origin_period and ultimate_change (float64). A draw’s total is the sum of its rows.

Methods

Name Description
to_polars() One of the result’s tables as a polars DataFrame.

to_polars()

One of the result’s tables as a polars DataFrame.

Usage

Source

to_polars(table="origins")

table is "origins" (the default), "totals", "quantiles" or "draws". Needs polars: pip install "ibnr[polars]".