predictive-modeling

name	predictive-modeling
description	Build an interpretable predictive model from a decision need through features, validation, calibration, error analysis, and deployment handoff.

Predictive Modeling

Use when this procedure is the primary professional method needed for the assignment.

Procedure

1. Confirm the decision or outcome this work must support, its scope, owner, constraints, and definition of success.
2. Establish the evidence baseline using labeled data, feature provenance, serving constraints, class balance, cost matrix, and subgroup definitions. Do not fill material gaps with assumptions when they can change the result.
3. Define target and leakage boundaries, establish baseline, split by real serving conditions, train candidates, inspect subgroup errors, calibrate, and document use limits.
4. Exercise realistic edge, failure, transition, or exception cases that could invalidate the result; record unresolved uncertainty explicitly.
5. Validate the output against the original outcome and any neighboring professional contracts so this skill does not silently absorb another specialist's authority.
6. Record the resulting artifact, measurements, decisions, provenance, and handoff information needed for another owner to reproduce or continue the work.

Quality gate

Performance is reproducible on representative holdout data and limitations are explicit enough to prevent misuse.