| EQ-01 // Markowitz Mean–Variance AI Spend Program (Explicit μ vs. Σ) |
minw wᵀΣw s.t. wᵀμ = μ*, wᵀ1 = 1, wᵀc ≤ BCFO μi = 𝔼[vi] (expected net value/task), Σij = Cov(vi,vj) = Cov(ei,ej) (if value = const − error) |
Unifies return and error notation rigorously: when net task value satisfies v_i = c_i − e_i with constant c_i (if value = const − error) and signed task error e_i, Cov(v_i, v_j) = Cov(e_i, e_j) minimizes joint failure variance at target net value μ*. |
| EQ-02 // Equal-Weight, Equal-Variance Correlated Fleet Floor |
Equal weights 1/N, equal σ2, pairwise ρ: Var(ē) = σ2 [ ρ + (1 − ρ) / N ] → σ2ρ |
States all three required assumptions explicitly: under equal weights, common variance σ², and pairwise correlation ρ, scaling N never drives risk to zero — variance hits the irreducible base-model floor σ²ρ. |
| EQ-03 // Effective Independent Headcount (N_eff — 'Correlation Is the New Headcount') |
Neff(N, ρ) = N / [ 1 + (N − 1)ρ ] (N=50, ρ=0.60 ⇒ Neff ≈ 1.64; ρ→1 ⇒ Neff→1) |
Exact diversification equivalence: 50 agents sharing a foundation model at ρ = 0.60 provide the statistical error reduction of only 1.64 independent agents (collapsing to 1.00 as ρ → 1). |
| EQ-04 // Systematic Base-Model Beta (β_i) vs. Idiosyncratic Agent Noise (CAPM) |
ei = βiFbase + εi ⇒ σp2(w) = (wᵀβ)2σbase2 + ∑i=1N wi2σε,i2 |
Prompt tweaks only reduce idiosyncratic σ²_{ε,i} (which diversifies as 1/N); systematic sensitivity β_i to shared foundation-model hallucinations requires orthogonal model families or deterministic invariants. |
| EQ-05 // Deterministic Code as Idealized Risk-Free Leg & Capital Market Line (Tobin Separation) |
σcode = 0 (idealized risk-free leg), μCML(σp) = μcode + [ (wtanᵀμ − μcode) / √(wtanᵀΣwtan) ] σp |
Rules, SQL reconciliations, and deterministic scripts serve as an idealized risk-free leg (σ_code = 0 on well-tested invariants); optimal CFO architecture blends deterministic code with the diversified multi-vendor LLM tangency portfolio w_tan along the CML. |
| EQ-06 // Finite-Sample Singularity (N > T) & Ledoit–Wolf Covariance Shrinkage |
rank(Semp) ≤ T − 1 < N (e.g. N=50, T=30d ⇒ det Semp = 0); Σ̂LW = (1 − δ*)Semp + δ*νIN |
With 50 agents and 30 days of evals, sample covariance S_emp is rank-deficient (rank ≤ 29) and non-invertible; Ledoit–Wolf shrinkage δ* restores positive-definiteness and out-of-sample frontier stability. |
| EQ-07 // Cowles Monograph 16 Downside Semivariance Matrix (Σ⁻, Markowitz 1959 Ch. 9) |
Σij−(τ) = 𝔼[ min(vi − τ, 0) · min(vj − τ, 0) ], Sp−(w) = wᵀΣ−(τ)w |
Penalizes only joint downside shortfalls below accounting accuracy target τ — aligning optimization with CFO asymmetry where upside speed is welcome but reporting misstatements are costly. |
| EQ-08 // Audit Materiality (M_audit) as Rockafellar–Uryasev CVaR Risk Budget |
CVaRα(L(w)) = minζ { ζ + (1−α)−1𝔼[max(L(w)−ζ, 0)] } ≤ Maudit |
Translates statutory SOX / external-audit materiality M_audit directly into an enforceable 99% tail-loss budget decomposed by Euler marginal risk contribution across every ledger workflow. |
| EQ-09 // Agent Sharpe Ratio & Bridge to Risk-Adjusted Agent ROE (Walrasian Equilibrium) |
SRi = ( 𝔼[Valuei − Costi] − μcode ) / σi ⇔ Walrasian Risk-Adjusted Agent ROE |
Measures excess net value per unit of outcome volatility σ_i over deterministic scripts — linking Markowitz/Sharpe portfolio weights directly to risk-adjusted Agent ROE in compute market equilibrium. |
| EQ-10 // Single-Model Banking Large-Exposure Cap (LEX) & Paired-Eval Multi-Vendor ρ̂ |
∑i∈m wi · EADi ≤ κLEXMaudit, ρ̂AB = (p11 − pApB) / √(pAqApBqB) |
Applies banking Large Exposures supervisory caps to any single foundation-model provider and verifies cross-vendor diversification (Anthropic, OpenAI, Google) empirically via paired benchmark evals. |
| EQ-11 // Non-Stationarity Across Model Releases & Covariance Rebalancing Trigger |
‖Σ̂t+1 − Σ̂t‖F / ‖Σ̂t‖F > γrebal ⇒ Re-estimate Σ̂t+1 & Rebalance wt+1* |
Every model version bump, quantization update, or prompt drift alters joint failure modes; continuous canary evals detect covariance regime shifts and rebalance agent weights automatically. |
| EQ-12 // Multi-Tier Cascade Spend, Selective HITL Gate & CFO RAROC Hurdle |
𝔼[ccasc] = c1 + ℙ(q1<τ1)c2 + …; Defer ⇔ perrLfin > wHtrev − cAI; RAROCAI ≥ rWACC |
Combines FrugalGPT/RouteLLM cost cascades (−65% to −90% token spend), economic Human-in-the-Loop exception gating, and CVaR-charged Return on Invested Capital. |