Stability

The module provides methods to perform stability analytics. It provides a table containing the Population Stability Index (PSI).

Accessor

Initialise the DataFrame with the stability method. Minimal working example:

df.crm.stability(score_ref="SCORE_REFERENCE", score="SCORE")

(
        df
        .crm.frequency(index="GRADE", column="DATE", cohort="ID")
        .table()
        .crm.stability(score_ref="SCORE_REF", score="SCORE")
        .table()
)

Parameters:

Name	Type	Description	Default
`score_ref`	`str`	Defines the reference population.	required
`score`	`Union[str, list]`	Defines the population to be compared to the reference population.	required

Returns:

Type	Description
`Stability`	Returns a class called "Stability" providing stability analytics methods.

Methods

`table(sort_by_col=None, sort_by_list=None, sort_asc=True, add_sum=False)`

Minimal working example:

df.crm.stability(score_ref="SCORE_REFERENCE", score="SCORE").table()

(
        df
        .crm.frequency(index="GRADE", column="DATE", cohort="ID")
        .table()
        .crm.stability(score_ref="SCORE_REF", score="SCORE")
        .table()
)

Parameters:

Name	Type	Description	Default
`sort_by_col`	`str`	Defines the column to sort.	`None`
`sort_by_list`	`list`	Defines the list to sort in case of categorical items which have no intrinsic sorting order.	`None`
`sort_asc`	`bool`	Sorts the previously defined column or list ascending.	`True`
`add_sum`	`bool`	Adds a sum row to the DataFrame.	`False`

Returns:

Type	Description
`DataFrame`	Returns the PSI.

Examples

data

>>> import credit_risk_modelling as crm
>>> data = crm.load_data.load_data()
>>> data

           DATE    ID GRADE  GRADE_PD OVERRIDE  OVERRIDE_PD  DEFAULT
0    2019-12-31    10     B    0.1000        B       0.1000        0
1    2019-12-31   100   BBB    0.0090       BB       0.0400        0
2    2019-12-31  1000   BBB    0.0090      BBB       0.0090        0
3    2019-12-31  1001   BBB    0.0090      BBB       0.0090        0
4    2019-12-31  1003   BBB    0.0090      BBB       0.0090        0
...         ...   ...   ...       ...      ...          ...      ...
4145 2023-12-31   994    AA    0.0010       AA       0.0010        0
4146 2023-12-31   995    AA    0.0010       AA       0.0010        0
4147 2023-12-31   996     A    0.0020        A       0.0020        0
4148 2023-12-31   998     B    0.1000        B       0.1000        0
4149 2023-12-31   999   AAA    0.0002      AAA       0.0002        0

[4150 rows x 7 columns]

.table()

>>> (
>>>     data
>>>     .loc[lambda df: df["DATE"].dt.year.isin([2022, 2023])]
>>>     .crm.frequency(index="GRADE", column="DATE", cohort="ID")
>>>     .table()
>>>     .crm.stability(score_ref="2023-12-31_PCT", score="2022-12-31_PCT")
>>>     .table(sort_by_list=crm.cfg.GRADES)
>>> )

  GRADE  2022-12-31_ABS  2022-12-31_PCT  2023-12-31_ABS  2023-12-31_PCT  PSI_2023-12-31_PCT-2022-12-31_PCT
0   AAA              63        0.076736              69        0.084044                           0.000665
1    AA              79        0.096224              95        0.115713                           0.003594
2     A             109        0.132765             106        0.129111                           0.000102
3   BBB             126        0.153471             106        0.129111                           0.004211
4    BB             135        0.164434             132        0.160780                           0.000082
5     B             107        0.130329             108        0.131547                           0.000011
6   CCC              94        0.114495              99        0.120585                           0.000316
7     D             108        0.131547             106        0.129111                           0.000046