reserving.OneYearFit
The simulated one-year view of every segment, from
Usage
reserving.OneYearFit()OdpBootstrap.one_year or MackBootstrap.one_year (model says which).
cdr is one joint distribution of the claims development result with the triangle’s keys and "origin" as dimensions, so cdr.aggregate(["lob"]) keeps the dependence between segments, and cdr.quantile(0.005) is minus the one-year value at risk at 99.5%. Per-origin lists run over the origins of each segment in turn, like the rows of to_frame() and the components of cdr. fitted, residuals and scale need a single-segment fit; for several segments use segment(...) or totals_frame(). fitted and scale are the ODP bootstrap’s (OdpBootstrapFit’s), mack is Mack’s bootstrap’s model (a MackFit of every segment, as chain_ladder), and residuals are either’s.
Attributes
| Name | Description |
|---|---|
| cdr | Joint distribution of the claims development result (opening less |
| chain_ladder | The bootstrap’s chain ladder: the factors the simulated next cells |
| development | Development ages in months. |
| fitted | The ODP bootstrap’s fitted incremental values, |
| index | Label of each segment, as Triangle.index. |
| keys | Names of the triangle’s key columns; empty without keys. |
| latest | Latest observed value per origin. |
| mack | Mack’s model behind MackBootstrap.one_year, with its lifetime |
| model |
The bootstrap’s model: "odp" or "mack".
|
| opening_reserve | The opening ultimate less the latest value, per origin. |
| opening_ultimate | The method’s ultimate per origin on the observed triangle. |
| origins | Origin period of each per-origin value and CDR component. |
| residuals |
The residuals the bootstrap resamples, [origin][development]:
|
| scale |
The ODP bootstrap’s scale parameter phi.
|
cdr
Joint distribution of the claims development result (opening less
cdr: PredictiveDistribution
closing ultimate) by segment and origin: the triangle’s keys and "origin" are its dimensions, one component per segment and origin, one row per simulation. Its mean, variance and quantile describe the total. Columns of draw_matrix() follow origins.
chain_ladder
The bootstrap’s chain ladder: the factors the simulated next cells
chain_ladder: ChainLadderFit
develop with (volume-weighted for the ODP, Mack’s averaging for Mack’s).
development
Development ages in months.
development: list[int]
fitted
The ODP bootstrap’s fitted incremental values,
fitted: list[list[float]]
[origin][development].
index
Label of each segment, as Triangle.index.
index: list[Any]
keys
Names of the triangle’s key columns; empty without keys.
keys: list[str]
latest
Latest observed value per origin.
latest: list[float]
mack
Mack’s model behind MackBootstrap.one_year, with its lifetime
mack: MackFit
standard errors and claims_development_result().
model
The bootstrap’s model: "odp" or "mack".
model: str
opening_reserve
The opening ultimate less the latest value, per origin.
opening_reserve: list[float]
opening_ultimate
The method’s ultimate per origin on the observed triangle.
opening_ultimate: list[float]
origins
Origin period of each per-origin value and CDR component.
origins: list[str]
residuals
The residuals the bootstrap resamples, [origin][development]:
residuals: list[list[float]]
the ODP’s adjusted Pearson residuals, as OdpBootstrapFit.residuals, or Mack’s scaled bias-adjusted residuals of the link ratios, [o][k] the link from age k to k + 1 (nan where there is none, from a zero, or behind a factor with a single link ratio).
scale
The ODP bootstrap’s scale parameter phi.
scale: float
Methods
| Name | Description |
|---|---|
| development_frame() | The bootstrap’s chain ladders’ development factors, one row per |
| segment() | The one-year view of one segment, chosen by key values as |
| to_frame() |
One row per segment and origin: the key columns, origin,
|
| totals_frame() | One row per segment: the key columns, the totals of to_frame()’s |
development_frame()
The bootstrap’s chain ladders’ development factors, one row per
Usage
development_frame()segment and age, as ChainLadderFit.development_frame. Needs pandas.
Returns
pandas.DataFrame
segment()
The one-year view of one segment, chosen by key values as
Usage
segment(**keys)ChainLadderFit.segment, with its part of the joint claims development result (same dimensions).
Returns
OneYearFit
to_frame()
One row per segment and origin: the key columns, origin,
Usage
to_frame()latest, the method’s opening_ultimate and opening_reserve, and the cdr_mean and cdr_std_dev of the simulated claims development result. Needs pandas.
Returns
pandas.DataFrame
totals_frame()
One row per segment: the key columns, the totals of to_frame()’s
Usage
totals_frame()columns, the ODP bootstrap’s scale (not for Mack’s), and the cdr_mean and cdr_std_dev of the segment’s total claims development result. Needs pandas.
Returns
pandas.DataFrame