Real Function is Concave iff its Negative is Convex
Jump to navigation
Jump to search
Theorem
Let $f$ be a real function.
Let $I \subseteq \R$ be an interval of $\R$.
Then $f$ is concave on $I$ if and only if $-f$ is convex on $\R$.
Proof
Necessary Condition
Let $f$ be concave on $I$.
Let $\alpha, \beta \in \R_{>0}$ such that $\alpha + \beta = 1$.
Then:
\(\ds \map f {\alpha x + \beta y}\) | \(\ge\) | \(\ds \alpha \map f x + \beta \map f y\) | Definition of Concave Real Function | |||||||||||
\(\ds \leadsto \ \ \) | \(\ds -\map f {\alpha x + \beta y}\) | \(\le\) | \(\ds -\paren {\alpha \map f x + \beta \map f y}\) | |||||||||||
\(\ds \) | \(=\) | \(\ds \alpha \paren {-\map f x} + \beta \paren {-\map f y}\) |
and so $-f$ is convex by definition.
$\Box$
Sufficient Condition
Let $-f$ be convex on $I$.
Let $\alpha, \beta \in \R_{>0}$ such that $\alpha + \beta = 1$.
Then:
\(\ds -\map f {\alpha x + \beta y}\) | \(\le\) | \(\ds \alpha \paren {-\map f x} + \beta \paren {-\map f y}\) | Definition of Convex Real Function | |||||||||||
\(\ds \leadsto \ \ \) | \(\ds \map f {\alpha x + \beta y}\) | \(\ge\) | \(\ds - \paren {\alpha \paren {-\map f x} + \beta \paren {-\map f y} }\) | |||||||||||
\(\ds \) | \(=\) | \(\ds \alpha \map f x + \beta \map f y\) |
and so $f$ is concave by definition.
$\blacksquare$
Sources
- 1977: K.G. Binmore: Mathematical Analysis: A Straightforward Approach ... (previous) ... (next): $\S 12.13$