Hölder Mean for Exponent 1 is Arithmetic Mean
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:
- $\map {M_1} {x_1, x_2, \ldots, x_n} = \dfrac {x_1 + x_2 + \cdots + x_n} n$
which is the arithmetic mean of $x_1, x_2, \ldots, x_n$.
Proof
Recall the definition of the Hölder mean with exponent $p$:
- $\ds \map {M_p} {x_1, x_2, \ldots, x_n} = \paren {\frac 1 n \sum_{k \mathop = 1}^n {x_k}^p}^{1 / p}$
Then:
\(\ds \map {M_1} {x_1, x_2, \ldots, x_n}\) | \(=\) | \(\ds \paren {\frac 1 n \sum_{k \mathop = 1}^n {x_k}^1}^{1 / 1}\) | ||||||||||||
\(\ds \) | \(=\) | \(\ds \frac 1 n \sum_{k \mathop = 1}^n {x_k}\) | simplifying |
which is the arithmetic mean by definition.
$\blacksquare$
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: $3.1.19$