reserving.OdpBootstrapFit
A fitted ODP bootstrap of every segment.
Usage
reserving.OdpBootstrapFit()reserves is one joint distribution with the triangle’s keys and "origin" as dimensions, so reserves.aggregate(["lob"]) keeps the dependence between segments. Per-origin lists run over the origins of each segment in turn, like the rows of to_frame() and the components of reserves. fitted, residuals and scale need a single-segment fit; for several segments use segment(...) or totals_frame(). fitted and residuals are nested lists indexed [origin][development], like one segment of Triangle.values, with nan where the triangle is not observed.
Attributes
| Name | Description |
|---|---|
| chain_ladder | The deterministic volume-weighted chain ladder the bootstrap is |
| development | Development ages in months. |
| fitted |
Fitted incremental values, [origin][development].
|
| index | Label of each segment, as Triangle.index. |
| keys | Names of the triangle’s key columns; empty without keys. |
| origins | Origin period of each per-origin value and reserve component. |
| reserves | Joint distribution of the reserve (the sum of future incremental |
| residuals |
Adjusted Pearson residuals (x - m) / sqrt(|m|) * sqrt(n / (n - p)),
|
| scale |
The scale parameter phi: the sum of squared unadjusted residuals
|
chain_ladder
The deterministic volume-weighted chain ladder the bootstrap is
chain_ladder: ChainLadderFit
centred on.
development
Development ages in months.
development: list[int]
fitted
Fitted incremental values, [origin][development].
fitted: list[list[float]]
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]
origins
Origin period of each per-origin value and reserve component.
origins: list[str]
reserves
Joint distribution of the reserve (the sum of future incremental
reserves: PredictiveDistribution
values) 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 and quantile describe the total reserve. Columns of draw_matrix() follow origins.
residuals
Adjusted Pearson residuals (x - m) / sqrt(|m|) * sqrt(n / (n - p)),
residuals: list[list[float]]
[origin][development]; nan where not observed or where the fitted value is zero.
scale
The scale parameter phi: the sum of squared unadjusted residuals
scale: float
over the degrees of freedom n - p.
Methods
| Name | Description |
|---|---|
| development_frame() | The chain ladders’ development factors, one row per segment and |
| segment() | The bootstrap of one segment, chosen by key values as |
| to_frame() |
One row per segment and origin: the key columns, origin, the
|
| totals_frame() | One row per segment: the key columns, the chain ladder’s totals, the |
development_frame()
The chain ladders’ development factors, one row per segment and
Usage
development_frame()age, as ChainLadderFit.development_frame. Needs pandas.
Returns
pandas.DataFrame
segment()
The bootstrap of one segment, chosen by key values as
Usage
segment(**keys)ChainLadderFit.segment, with its part of the joint reserves (same dimensions).
Returns
OdpBootstrapFit
to_frame()
One row per segment and origin: the key columns, origin, the
Usage
to_frame()chain ladder’s latest, ultimate and reserve, and the mean and std_dev of the bootstrapped reserve. Needs pandas.
Returns
pandas.DataFrame
totals_frame()
One row per segment: the key columns, the chain ladder’s totals, the
Usage
totals_frame()Returns
pandas.DataFrame