Portfolio Optimization¶
Choose a feasible combination of investments or projects by optimizing a declared portfolio-level criterion.
Core Idea¶
Portfolio optimization chooses a combination of investments or projects by comparing feasible combinations under a declared portfolio-level criterion. Decisions may be asset weights or binary project selections. The value or risk of the combination can differ from a sum of independent member scores.
Scope of Application¶
Financial mean-variance allocation is one formulation: it weighs expected return against variance, including covariances among securities. R&D project portfolios can instead use binary project decisions, value, risk, interdependencies and funding or staffing limits.
Clarity¶
A held portfolio is not automatically an optimized one. The candidates, decision variables, feasible set, outcome model and choice criterion must be named before calling a solution optimal. The efficient frontier belongs to particular risk-return formulations, not every portfolio problem.
Manages Complexity¶
The approach compares combinations under joint consequences and constraints instead of ranking members separately. Modeling cross-member effects can overturn an additive ranking, but requires more pairwise or project-interaction estimates whose reliability must be tested. The result remains conditional on uncertain returns, values, risks and interactions.
Abstract Reasoning¶
Two securities with imperfectly correlated returns can form a mixture with different variance from either holding alone. Two R&D projects can also interact or compete for scarce staff. In both cases, the best combination cannot be inferred solely from the highest individual score.
Knowledge Transfer¶
The candidate–combination–constraint–objective structure transfers from finance to project selection. Covariance-based variance, market return and a continuous weight model need not transfer with it.
This is a strict kind of Optimization Problem: portfolio membership or weights narrow the parent’s feasible-choice-and-objective structure.
Relationships to Other Abstractions¶
Current abstraction Portfolio Optimization Domain-specific
Parents (1) — more general patterns this builds on
-
Portfolio Optimization is a kind of Optimization Problem Domain-specific
Portfolio optimization chooses among feasible portfolio decisions by a specified objective.
Hierarchy path (1) — routes to 1 parentless root
- Portfolio Optimization → Optimization Problem → Optimization
Neighborhood in Abstraction Space¶
Portfolio Optimization sits in a moderately populated region (53rd percentile for distinctiveness): it has near-neighbors but no dense thicket of look-alikes.
Family — Strategic Decision Biases & Mechanisms (29 abstractions)
Nearest neighbors
- Portfolio (finance) — 0.88
- Duck Typing — 0.85
- Public Goods Game — 0.85
- Identifiable Victim Effect — 0.85
- Growth–share matrix — 0.85
Computed from structural-signature embeddings · 2026-10-08