A formal framework for capital allocation across Africa's innovation economy - grounded in stochastic modeling, sectoral signal processing, and multi-horizon risk architecture.
This paper defines a unified valuation and risk methodology for Sidakye Capital's multi-strategy private portfolio. It spans venture and startup equity, infrastructure and project-finance assets, and structured resource-linked positions under one consistent measurement framework.
The thesis aligns valuation with market-participant exit-price logic and emphasizes scenario-consistent cash-flow modeling, transparent discount-rate construction, and disciplined governance for Level 3 measurements. The same system is used to produce both fair-value marks and decision-ready risk signals for hedging, leverage, and portfolio construction.
| Symbol | Meaning |
|---|---|
| \(CF_{i,s}(u_k)\) | Scenario-dependent cash flow for position \(i\) at time \(u_k\) |
| \(p_s\) | Committee-governed probability weight for scenario \(s\) |
| \(r_{i,s}\) | Scenario-consistent discount rate or WACC for position \(i\) |
| \(Adj_i(t)\) | Policy-governed overlays (liquidity, execution, conservatism) |
The methodology is tested through pricing-error back-testing, tail-risk calibration, and hedge-effectiveness diagnostics so outperformance claims remain empirical rather than narrative.
The framework is built on six principles: scenario consistency over point estimates, maximal use of observable inputs, explicit treatment of unobservables, model-and-judgment separation, clean aggregation integrity, and full governance traceability.
"Scenario-consistent assumptions outperform narrative marks in private markets."
A macro-structural overlay (PWRO) links International Futures domain shocks to valuation assumptions through documented relevance weights and committee-approved coefficients. This creates a controlled bridge between long-horizon regime risk and present valuation parameters.
For asset \(a\), \(R_a\) is the scenario-weighted dispersion of structural impacts across IFs modules. It is used as a governed scalar that deforms discount rates, scenario probabilities, and selected cash-flow assumptions.
This theory layer keeps growth, policy, infrastructure, and financing assumptions internally coherent while preserving auditable links to evidence and committee decisions.
Implementation follows a fixed pipeline: classify each instrument, define unit-of-account and currency boundary, build scenario cash flows or payoff functions, calibrate discounting and overlays, then produce valuation and sensitivity outputs under governed review.
Positions are segmented into VSU (venture/startups), IF (infrastructure/project assets), and DRF (de-risk/structured exposures). Each bucket has a default valuation engine and a required cross-check method so reporting remains comparable across heterogeneous assets.
| Symbol | Description |
|---|---|
| \(PV_s\) | Present value under scenario \(s\) |
| \(\Pr_g\) | Stage-gate success probability for gate \(g\) |
| \(NPV_s\) | Net present value of successful execution path in scenario \(s\) |
| \(SV_s\) | Salvage or downside continuation value in scenario \(s\) |
For stage-dependent assets, value is decomposed into success and non-success branches with explicit gate probabilities, then weighted by scenario likelihoods. This avoids overstating upside in execution-heavy positions.
The final output package includes position value, scenario contribution bridge, sensitivity map, evidence links, and reviewer sign-off metadata for committee approval.
Risk is measured with two synchronized lenses: scenario distributions for non-linear execution outcomes, and factor sensitivities for hedgeable macro exposures such as FX, commodities, rates, and demand.
Tail governance uses quantile and expected-shortfall metrics together. VaR gives breach frequency expectations, while CVaR captures the average severity inside the tail.
Deterministic stress tests then revalue the portfolio under explicit shocks (for example FX minus 20 percent, commodity minus 30 percent, and rates plus 200 bps) to produce ranked NAV-impact diagnostics and breach triggers.
Capital allocation is solved as a constrained optimization over the tri-layer structure. Expected return is balanced against covariance risk, concentration, and liquidity constraints while preserving full-investment consistency.
Allocation decisions are paired with hedge sizing and risk-budgeted leverage ceilings so exposure growth is always constrained by downside capacity.
The committee pack receives recommended bucket weights, hedge notionals, expected variance and tail reduction, and any leverage breaches with required actions.
The thesis establishes a single, governed valuation language across venture, infrastructure, and structured private assets. It links fair-value marks, risk diagnostics, hedge design, and leverage policy into one reproducible decision framework.
Method superiority is tested through event-based pricing error (MAE/RMSE), mark timeliness, VaR exceedance alignment, and realized hedge effectiveness. This keeps performance claims falsifiable and continuously improvable.
"Capital allocation is strongest when valuation discipline, risk measurement, and governance evidence move together."
Execution follows a phased rollout: (1) data and governance foundation, (2) bucket-specific valuation engines, (3) risk and hedging integration, (4) back-testing and calibration, and (5) continuous reporting refinement.
This document contains forward-looking statements and mathematical projections for informational purposes only. All models and projections are subject to assumptions and estimation error. Not investment advice. For qualified institutional investors only. See full disclosure.