Credit Scoring Module Class: 2022 & 2023

---

Overview

• Odds based Grouping (OBG)

  • OBGEncoder

    • pred_var: Name of predictor variable. Values can be either continuous or categorical.
    • target_var: Name of binary target variable.

    .fit

    • df: DataFrame containing pred_var and target_var.
    • max_delta: max difference between odds for merging two levels. default: 0.05
    • min_bins: minimum number of bins. default: 3
    • q: number of quantiles when converting continuous variable to categorical. default: 10

    .transform

    • df: Transform pred_var based on fitted bins.
    • impute: Boolean indicating whether to impute missing values. default: False
    • impute_value: Category level to impute missing values with. default: 'Missing' or 'nan'

    .fit_transform

    • df: DataFrame containing pred_var and target_var. Transform pred_var based on fitted bins.

    fit_dict: dictionary containing the matched category levels and fitted bins.

    lookup: dictionary containing cutoff values for continuous variable (empty if pred_var is categorical).

• Weight of Evidence (WOE)

  • WOEEncoder

    • pred_var: Name of predictor variable.Values can be either continuous or categorical.
    • target_var: Name of binary target variable.
    • target_value: Value indicating event. default: 1.

    .fit

    • df: DataFrame containing pred_var and target_var.
    • stop_limit: Stops WOE based merging of the predictor's classes/levels in case the resulting information value (IV) decreases more than (e.g. 0.05 = 5%) compared to the preceding binning step. stop_limit=0 will skip any WOE based merging. Increasing the stop_limit will simplify the binning solution and may avoid overfitting. Accepted value range: 0 to 0.5. default: 0.1.
    • q: number of quantiles when converting continuous variable to categorical. default: 10

    .transform

    • df: Transform pred_var based on fitted bins
    • impute: Boolean indicating whether to impute missing values. default: False
    • impute_value: Category level to impute missing values with. default: 'Missing' or 'nan'

    .fit_transform

    • df: DataFrame containing pred_var and target_var. Transform pred_var based on fitted bins.

    .test_limit

    • df: DataFrame containing pred_var and target_var to test stop limits at 1%, 2.5%, 5% and 10%.

    fit_dict: dictionary containing the matched category levels and fitted bins.

    lookup: dictionary containing cutoff values for continuous variable (empty if pred_var is categorical).