Definition
Constrained optimization chooses a decision that optimizes an objective while satisfying equalities, inequalities or bounds. In portfolio work, constraints can encode budget, exposure, leverage or position limits. A constraint can restrict a feasible problem and sometimes make an unbounded objective attain a minimum, but it can also change the estimand being optimized.
How it is calculated
Minimize f(x) subject to every inequality g_j(x) less than or equal to zero and equality h_k(x) equal to zero.
| Symbol | Meaning | Unit |
|---|---|---|
| x | decision vector | decision-specific |
| f | objective | objective-specific |
| g | inequality constraint | constraint-specific |
| h | equality constraint | constraint-specific |
Interpretation
The objective says what to improve, and constraints say what is allowed. The solution is the best point inside that permitted region, not necessarily the unconstrained best point. Tight limits can dominate the answer and should be reported.
Assumptions and conventions
-
The feasible set and constraint units are well defined.
-
The optimization method handles the problem's convexity or nonconvexity correctly.
Each constraint compares quantities in compatible units; x and f retain their problem-specific units.
Limitations and common misreadings
-
A constraint is not merely a post-processing filter if it changes the optimization problem.
-
Feasibility does not imply that the objective is well estimated.
References
- Stephen Boyd, Lieven Vandenberghe. (2004). Convex Optimization. Cambridge University Press. Section 4.1, Optimization problems. doi:10.1017/CBO9780511804441. Source

