# Example Python code for implementing a credit risk logistic regression scorecard
# with synthetic data generation for demonstration purposes
import pandas as pd
import numpy as np
from sklearn.linear_model import LogisticRegression
from sklearn.model_selection import train_test_split
from sklearn.metrics import roc_auc_score

# Generate synthetic credit dataset
np.random.seed(42)
n_samples = 1000

# Features: example numeric features representing financial attributes
data = pd.DataFrame({
    'age': np.random.randint(18, 70, size=n_samples),
    'income': np.random.normal(50000, 15000, size=n_samples),
    'credit_score': np.random.normal(650, 50, size=n_samples),
    'debt_to_income': np.random.uniform(0, 1, size=n_samples),
    'num_of_loans': np.random.randint(0, 10, size=n_samples)
})

# Target variable: default_flag, probabilistically generated based on features with noise
logit = (
    -0.05 * data['age'] +
    -0.00002 * data['income'] +
    -0.01 * data['credit_score'] +
    2.5 * data['debt_to_income'] +
    0.1 * data['num_of_loans'] -
    1.0
)
prob = 1 / (1 + np.exp(-logit))
data['default_flag'] = np.random.binomial(1, prob)

# Prepare features and target
X = data.drop('default_flag', axis=1)
y = data['default_flag']

# Split data into train and validation sets
train_x, val_x, train_y, val_y = train_test_split(X, y, test_size=0.3, random_state=42)

# Initialize logistic regression model
model = LogisticRegression(max_iter=1000)

# Fit model on training data
model.fit(train_x, train_y)

# Predict probabilities on validation data
val_preds = model.predict_proba(val_x)[:, 1]

# Calculate AUC as validation metric
auc_score = roc_auc_score(val_y, val_preds)
print(f'Validation AUC: {auc_score:.4f}')