Forecasting • Pricing • Decision Theory
From data → scenarios → optimal action.
SKU-level forecasting focused on bias, stability, and operational risk.
Models are evaluated per SKU using a bias-aware score (MAE + |Bias|) to identify signals that remain directionally stable under volatility.
Risk-aware price selection under demand uncertainty.
Candidate prices are evaluated using profit distributions, leverage, and downside exposure to return deploy, hold, or no-go decisions.
A minimal, auditable decision engine for choosing actions under uncertainty.
Actions are evaluated using expected loss, minimax regret, and CVaR to produce a single, defensible recommendation with explicit assumptions.
If you’re exploring forecasting stability, pricing under uncertainty,
or decision governance, you can reach me here.
v0.1 — public decision-grade demo (Dec 2025)