bayesline.api.equity.InlineUploadedExposureGroupSettings#

model InlineUploadedExposureGroupSettings#

Exposure group settings for exposures taken directly from an uploaded dataset.

Reads one factor group of an uploaded exposures dataset at report time, without requiring a risk dataset rebuild. Assets in the upload that are not part of the risk dataset are dropped with a warning when the report is built.

Uploaded exposures are stored as 16 bit floats, so the exposures used here are the 16 bit round trip of the uploaded values.

Fields#
Field	Type	Required	Default	Constraints
`exclude`	`list`[`str`]	No	factory
`exposure_source`	`str`	Yes		pattern=``^(?:[^/s]+/)?[A-Za-z][A-Za-z0-9_-]*$``, min_length=1
`exposure_type`	`Literal`[‘inline_uploaded’]	No	`'inline_uploaded'`
`factor_group`	`str`	Yes		pattern=``^[A-Za-z][A-Za-z0-9_-]*$``, min_length=1
`fill_miss`	`bool`	No	`False`
`forward_fill`	`bool`	No	`False`
`gaussianize`	`bool`	No	`False`
`gaussianize_maintain_zeros`	`bool`	No	`False`
`include`	`Literal`[‘All’] \| `list`[`str`]	No	`'All'`
`net_of`	`list`[`str`]	No	factory
`standardize_method`	`Literal`[‘none’, ‘equal_weighted’]	No	`'none'`

field exclude: list[str] [Optional]#: Factor names to exclude from the uploaded dataset.

field exposure_source: str [Required]#

The name of the uploaded exposures dataset.

Constraints:

pattern = ^(?:[^/\s]+/)?[A-Za-z][A-Za-z0-9_-]*$
min_length = 1

field exposure_type: Literal['inline_uploaded'] [Optional]#

field factor_group: str [Required]#

The factor group in the uploaded dataset to take exposures from.

Constraints:

pattern = ^[A-Za-z][A-Za-z0-9_-]*$
min_length = 1

field fill_miss: bool [Optional]#: Whether to fill in missing exposures.

field forward_fill: bool [Optional]#: Whether to forward-fill the exposures along time, gated by the priced-day mask (no fill on days the asset is not in the modeling universe). Applied before gaussianizing and filling missing values.

field gaussianize: bool [Optional]#: Whether to gaussianize the exposures.

field gaussianize_maintain_zeros: bool [Optional]#: Whether to maintain zeros when gaussianizing the exposures. This only takes effect during gaussianization, so it requires gaussianize.

field include: Literal['All'] | list[str] [Optional]#: Factor names to include from the uploaded dataset. ‘All’ includes all.

field net_of: list[str] [Optional]#: Factor groups this group’s exposures are taken net of (weighted-orthogonalized against) under the exposure settings’ orthogonalization_weights, e.g. a style block net of [‘market’, ‘industry’]. See orthogonalization_weights for what moves in a consuming model and the covariance caveat.

field standardize_method: Literal['none', 'equal_weighted'] [Optional]#: The method to use for standardizing the exposures. If ‘none’, no standardization is applied. If ‘equal_weighted’, then the exposures are standardized with the mean and standard deviation of the estimation universe.