Canonical: https://knowlpedia.com/convex-analysis/sums-and-scalar-multiples-of-convex-sets-are-convex/

# Sums and scalar multiples of convex sets are convex

Minkowski sums and dilations preserve convexity

**Proposition.**
Let \(\Omega_1,\Omega_2\) be [convex](https://knowlpedia.com/convex-analysis/convex-set/) subsets of a real vector space \(X\) and let \(\lambda\in\mathbb{R}\). Define the [Minkowski sum and scalar multiple](https://knowlpedia.com/convex-analysis/set-operations-sum-scalar-multiple-difference/)
\[
\Omega_1+\Omega_2:=\{x_1+x_2:x_1\in\Omega_1,\ x_2\in\Omega_2\},\qquad
\lambda\Omega_1:=\{\lambda x:x\in\Omega_1\}.
\]

Then \(\Omega_1+\Omega_2\) and \(\lambda\Omega_1\) are convex.

## Remarks

**Context.** This is a key stability property: adding convex "uncertainty sets" or scaling constraints preserves convexity.

**Proof sketch.** For sums, take \(u_i=x_i+y_i\in\Omega_1+\Omega_2\) and use convexity of each set to show \(\lambda u_1+(1-\lambda)u_2\) decomposes as a sum of two points in \(\Omega_1\) and \(\Omega_2\). For scalar multiples, note \(\lambda(\theta x+(1-\theta)y)=\theta(\lambda x)+(1-\theta)(\lambda y)\).
