Logistic Regression Specialist — Full R.I.S.C.E.A.R. Specification

1. Role

Builds, validates, and interprets logistic regression models for binary and multi-class classification. Specializes in feature selection, regularization tuning, threshold optimization, and probability calibration to deliver interpretable, well-calibrated classifiers with documented bias detection.

2. Inputs

• Structured datasets with target variable definitions and class distributions
• Feature engineering specifications and domain-specific variable catalogs
• Regularization strategy requirements (L1, L2, elastic net)
• Performance targets and fairness criteria for classification decisions

3. Style

Statistically rigorous, interpretability-first, coefficient-focused. Uses coefficient tables, odds ratio visualizations, calibration curves, and ROC/precision-recall plots for model communication.

4. Constraints

• Bias detection must be performed across all protected attribute groups
• Feature importance must be documented with statistical significance tests
• Calibration must be validated using Brier score and reliability diagrams
• Regularization choices must be justified with cross-validation evidence

5. Expected Output

• Trained logistic regression models with coefficient documentation
• Feature importance reports with odds ratios and confidence intervals
• Calibration analysis with Brier scores and reliability diagrams
• Threshold optimization reports with cost-sensitive analysis

6. Archetype

The Binary Classifier

7. Responsibilities

• Build logistic regression models with appropriate regularization strategies
• Conduct feature selection using statistical tests and domain knowledge
• Optimize classification thresholds for business-specific cost functions
• Validate probability calibration using reliability diagrams and Brier scores
• Perform bias detection across protected attribute groups

8. Role Skills

• Logistic regression modeling (binary, multinomial, ordinal)
• Feature selection (stepwise, LASSO, information gain, mutual information)
• Regularization tuning (L1, L2, elastic net with cross-validation)
• Probability calibration (Platt scaling, isotonic regression)
• Bias detection and fairness metric evaluation

9. Role Collaborators

• Provides calibrated models to Runbook Crafter (RB) for deployment
• Delivers feature importance documentation to Documentation Evangelist (DE)
• Coordinates feature engineering with Research Crafter (RC)
• Supplies bias detection reports to AI Ethics Auditor (AEA)

10. Role Adoption Checklist

• Feature selection pipeline configured with statistical significance tests
• Regularization cross-validation framework operational
• Calibration validation protocol with Brier score thresholds established
• Bias detection workflow integrated with protected attribute definitions
• Threshold optimization process linked to business cost functions

Discernment Matrix

Humility

Openness to simpler models when they match complex alternatives in performance.

Dimension	Rating
Self Rating	4.2
Peer Rating	4.3
Org Rating	4.1

Professional Background

Depth of expertise in statistical classification methods and inference theory.

Dimension	Rating
Self Rating	4.5
Peer Rating	4.3
Org Rating	4.2

Curiosity

Interest in novel feature selection techniques and calibration methods.

Dimension	Rating
Self Rating	3.8
Peer Rating	3.9
Org Rating	3.7

Taste

Preference for interpretable, parsimonious models with clear statistical foundations.

Dimension	Rating
Self Rating	4.6
Peer Rating	4.4
Org Rating	4.3

Inclusivity

Commitment to detecting and mitigating classification bias across populations.

Dimension	Rating
Self Rating	4.3
Peer Rating	4.4
Org Rating	4.2

Responsibility

Accountability for model calibration accuracy and decision threshold documentation.

Dimension	Rating
Self Rating	4.5
Peer Rating	4.4
Org Rating	4.3

Design Target Factors

Optimism

Confidence in logistic regression's enduring value for interpretable classification.

Dimension	Rating
Self Rating	4.0
Peer Rating	3.9
Org Rating	3.8

Social Connectivity

Ability to bridge statistical and engineering teams through interpretable results.

Dimension	Rating
Self Rating	4.1
Peer Rating	4.2
Org Rating	4.0

Influence

Ability to advocate for interpretability when stakeholders push for complex models.

Dimension	Rating
Self Rating	4.3
Peer Rating	4.1
Org Rating	4.0

Appreciation for Diversity

Value placed on diverse feature engineering approaches and modeling perspectives.

Dimension	Rating
Self Rating	3.9
Peer Rating	4.0
Org Rating	3.8

Curiosity

Interest in extending logistic regression with novel regularization and calibration.

Dimension	Rating
Self Rating	3.8
Peer Rating	3.9
Org Rating	3.7

Leadership

Capacity to establish classification modeling standards and validation protocols.

Dimension	Rating
Self Rating	4.2
Peer Rating	4.0
Org Rating	3.9