Proof that Optimal Dual Variables Represent Shadow Prices

Exploring the cinematic intuition of Proof that Optimal Dual Variables Represent Shadow Prices.

Visualizing...

Our institutional research engineers are currently mapping the formal proof for Proof that Optimal Dual Variables Represent Shadow Prices.

Apply for Institutional Early Access →

The Formal Theorem

Consider the primal linear program: Minimize z=cTx z = c^T x subject to Ax=b Ax = b and x0 x \ge 0 . Let x x^* be an optimal solution with associated optimal dual variables y y^* . The shadow price of the i i -th resource constraint, defined as the sensitivity of the optimal objective value z z^* to marginal changes in the resource vector bi b_i , is given by the i i -th component of the optimal dual vector:
zbi=yi \frac{\partial z^*}{\partial b_i} = y^*_i

Analytical Intuition.

Imagine the primal problem as an industrial furnace consuming raw materials b b to generate profit z z . We are locked in a cage of constraints, seeking the most efficient pathway. When we introduce the Dual, we shift our perspective from the resources themselves to the 'value' of the constraints governing them. The optimal dual variable yi y^*_i acts as a internal valuation mechanism. If we increment the availability of resource bi b_i by a microscopic unit ϵ \epsilon , the optimal objective z z^* shifts proportionally to yi y^*_i . It is the 'rate of exchange' between the capacity constraint and the objective function. Mathematically, this arises from the sensitivity analysis of the optimal basis B B , where the change in the optimal objective is Δz=cBTB1Δb \Delta z = c_B^T B^{-1} \Delta b . Since yT=cBTB1 y^{*T} = c_B^T B^{-1} , the dual variable captures exactly how much a marginal unit of resource contributes to your bottom line, essentially telling you: 'If you could acquire one more unit of this constraint, your profit would increase by exactly this much.'
CAUTION

Institutional Warning.

Students often mistake dual variables for fixed costs. They are not static; they are marginal values tied to the current optimal basis. If a resource is not fully utilized (slack exists), the shadow price yi y^*_i must be zero due to Complementary Slackness, as extra resource provides no marginal gain.

Academic Inquiries.

01

Why does the shadow price become zero when there is slack in the constraint?

By Complementary Slackness, if a constraint is not binding at the optimum, the corresponding dual variable must be zero. This aligns with economic intuition: if you have leftover resources, an additional unit of that resource has no marginal value.

02

Does the shadow price remain valid for large changes in b?

No. Shadow prices are local derivatives. As b b changes significantly, the optimal basis B B may change, resulting in a piecewise linear, convex objective function where shadow prices shift at basis transition points.

Standardized References.

  • Definitive Institutional SourceBertsimas, D., & Tsitsiklis, J. N., Introduction to Linear Optimization

Institutional Citation

Reference this proof in your academic research or publications.

NICEFA Visual Mathematics. (2026). Proof that Optimal Dual Variables Represent Shadow Prices: Visual Proof & Intuition. Retrieved from https://nicefa.org/library/linear-and-integer-programming/proof-that-optimal-dual-variables-represent-shadow-prices

Dominate the Logic.

"Abstract theory is just a movement we haven't seen yet."