Limit of Hölder Mean as Exponent tends to Negative Infinity
Jump to navigation
Jump to search
Theorem
Let $x_1, x_2, \ldots, x_n \in \R_{\ge 0}$ be real numbers.
For $p \in \R_{\ne 0}$, let $\map {M_p} {x_1, x_2, \ldots, x_n}$ denote the Hölder mean with exponent $p$ of $x_1, x_2, \ldots, x_n$.
Then:
- $\ds \lim_{p \mathop \to -\infty} \map {M_p} {x_1, x_2, \ldots, x_n} = \min \set {x_1, x_2, \ldots, x_n}$
Proof
Let $p \in \R$ such that $p \ne 0$.
Let it be assumed (or arranged) that:
- $x_1 \ge x_2 \ge \cdots \ge x_n$
Then:
\(\ds \lim_{p \mathop \to -\infty} \map {M_p} {x_1, x_2, \ldots, x_n}\) | \(=\) | \(\ds \dfrac 1 {\ds \lim_{p \mathop \to +\infty} \map {M_p} {\dfrac 1 {x_1}, \dfrac 1 {x_2}, \ldots, \dfrac 1 {x_n} } }\) |
|
|||||||||||||
\(\ds \) | \(=\) | \(\ds \dfrac 1 {\paren {\dfrac 1 {x_n} } }\) | Limit of Hölder Mean as Exponent tends to Infinity | |||||||||||||
\(\ds \) | \(=\) | \(\ds x_n\) | ||||||||||||||
\(\ds \) | \(=\) | \(\ds \min \set {x_1, x_2, \ldots, x_n}\) | by hypothesis: $x_1 \ge x_2 \ge \cdots \ge x_n$ |
$\blacksquare$
Sources
- 1964: Milton Abramowitz and Irene A. Stegun: Handbook of Mathematical Functions ... (previous) ... (next): $3$: Elementary Analytic Methods: $3.1$ Binomial Theorem etc.: Generalized Mean: $3.1.17$