Arithmetic, Geometric, Harmonic Means: Definition and Key Theorems

Arithmetic, Geometric, Harmonic Means

Arithmetic, Geometric, Harmonic Means are fundamental concepts in Mathematics used extensively in various domains. These means represent ways to summarize sets of numbers and hold significant importance in analytical studies, finance, and statistics.

What You Will Learn?

  • Definition: Arithmetic Mean
  • Definition: Geometric Mean
  • Definition: Harmonic Mean
  • Theorem-1: \( \pmb{A.M.\geq G.M.} \)
  • Theorem-2: \( \pmb{G.M.\geq H.M.} \)

Things to Remember

Before diving into this post, make sure you are familiar with: Basic Definitions and Concepts of
  1. Inequlities

Arithmetic, Geometric, Harmonic Means

  Arithmetic, Geometric, Harmonic Means have been integral to Mathematics since ancient times. Their roots can be traced to ancient Greek mathematicians, where these means were studied to understand ratios and proportions in geometry. These concepts play a vital role in modern-day problem-solving, offering techniques to compare and analyze numerical data effectively. Understanding these means aids in solving real-world problems ranging from engineering to economics.

Arithmetic Mean

  Definition:
  Let \(\pmb{a_{1},a_{2},…,a_{n}} \) be \(\pmb{n} \) positive real numbers. Then the Arithmetic Mean (A.M) of \(\pmb{a_{1},a_{2},…,a_{n}} \) is defined by \(\pmb{\frac{a_{1}+a_{2}+…+a_{n}}{n} } \).

Geometric Mean

  Definition:
  Let \(\pmb{a_{1},a_{2},…,a_{n}} \) be \(\pmb{n} \) positive real numbers. Then the Geometric Mean (G.M) of \(\pmb{a_{1},a_{2},…,a_{n}} \) is defined by \(\pmb{\sqrt[n]{a_{1}a_{2}…a_{n}} } \).

Harmonic Mean

  Definition:
  Let \(\pmb{a_{1},a_{2},…,a_{n}} \) be \(\pmb{n} \) positive real numbers. Then the Harmonic Mean (H.M) of \(\pmb{a_{1},a_{2},…,a_{n}} \) is defined by \(\pmb{\frac{n}{\frac{1}{a_{1}}+\frac{1}{a_{2}}+…+\frac{1}{a_{n}}} } \).

Theorem-1: \( \pmb{A.M.\geq G.M.} \)

  Statement:

  Let \(\pmb{a_{1},a_{2},…,a_{n}} \) be \(\pmb{n} \) positive real numbers then \(\pmb{ \frac{a_{1}+a_{2}+…+a_{n}}{n}\geq \sqrt[n]{a_{1}a_{2}…a_{n}} }\). The equality occurs when \(\pmb{a_{1}=a_{2}=…=a_{n}} \).


  Proof:
  Let \(\pmb{a_{1},a_{2},…,a_{n}} \) be \(\pmb{n} \) positive real numbers.

  • Case-1: Let \(\pmb{n=2^{k} }\) where \(\pmb{k }\) is positive integers.
    For any two positive real numbers \(\pmb{p,q }\)
    \begin{align} &\pmb{pq=\left(\frac{p+q}{2} \right)^{2}-\left(\frac{p-q}{2} \right)^{2}}\nonumber\\ \implies & \pmb{pq \leq \left(\frac{p+q}{2} \right)^{2}} \end{align} Since \(\pmb{ \left(p-q\right)^{2}\geq 0 } \). The equalily occurs if \(\pmb{ p=q } \).
    Putting \( \pmb{ p=a_{1},~q=a_{2}} \) in (1), we have \begin{align} &\pmb{a_{1}a_{2} \leq \left(\frac{a_{1}+a_{2}}{2} \right)^{2}} \\ \implies & \pmb{ \frac{a_{1}+a_{2}}{2}\geq \sqrt[2]{a_{1}a_{2}} } \end{align} The equality occurs when \(\pmb{a_{1}=a_{2}} \).
    Therefore the statement is true for \(\pmb{n=2 }\).
    Putting \( \pmb{ p=a_{3},~q=a_{4}} \) in (1), we have \begin{align} \pmb{a_{3}a_{4} \leq \left(\frac{a_{3}+a_{4}}{2} \right)^{2}} \end{align} Multiplying (2) and (3), we get \begin{align} &\pmb{a_{1}a_{2}a_{3}a_{4} \leq \left(\frac{a_{1}+a_{2}}{2} \right)^{2}\left(\frac{a_{3}+a_{4}}{2} \right)^{2}} \nonumber\\ \implies & \pmb{a_{1}a_{2}a_{3}a_{4} \leq \left[ \left(\frac{a_{1}+a_{2}}{2} \right)\left(\frac{a_{3}+a_{4}}{2} \right)\right]^{2}} \end{align} Putting \( \pmb{ p=\frac{a_{1}+a_{2}}{2} ,~q=\frac{a_{3}+a_{4}}{2}} \) in (1), we get \begin{align} &\pmb{\left(\frac{a_{1}+a_{2}}{2} \right)\left(\frac{a_{3}+a_{4}}{2} \right) \leq \left[\frac{\left(\frac{a_{1}+a_{2}}{2} \right)+\left(\frac{a_{3}+a_{4}}{2} \right)}{2} \right]^{2}} \nonumber\\ \implies & \pmb{\left[\left(\frac{a_{1}+a_{2}}{2} \right)\left(\frac{a_{3}+a_{4}}{2} \right)\right]^{2} \leq \left[\frac{a_{1}+a_{2}+a_{3}+a_{4} }{4} \right]^{4}} \end{align} From (4) and (5), we have \begin{align} & \pmb{a_{1}a_{2}a_{3}a_{4} \leq \left[\frac{a_{1}+a_{2}+a_{3}+a_{4} }{4} \right]^{4} }\nonumber\\ \implies & \pmb{ \frac{a_{1}+a_{2}+a_{3}+a_{4} }{4}\geq \sqrt[4]{a_{1}a_{2}a_{3}a_{4}} } \end{align} The equality occurs when \(\pmb{a_{1}=a_{2}=a_{3}=a_{4}} \).
    Therefore the statement is true for \(\pmb{n=2^{2}=4 }\).
    Continuing this process, after \(\pmb{ k }\) steps, we have for \(\pmb{n=2^{k} }\) \begin{align} \pmb{ \frac{a_{1}+a_{2}+…+a_{n}}{n}\geq \sqrt[n]{a_{1}a_{2}…a_{n}} } \end{align} The equality occurs when \(\pmb{a_{1}=a_{2}=…=a_{n}} \).

  • Case-2: Let \(\pmb{n\ne 2^{k} }\).
    Then \(\pmb{2^{k-1} \lt n \lt 2^{k} }\). Let \(\pmb{ 2^{k}=n+t }\) for some positive integers \(\pmb{t} \).
    Let \begin{align} \pmb{ m=\frac{a_{1}+a_{2}+…+a_{n}}{n} } \end{align} Consider the \(\pmb{n+t} \) positive real numbers \(\pmb{a_{1},a_{2},…,a_{n}} \) and \(\pmb{m,m,…} \) \(\pmb{t} \)-times
    we have for \(\pmb{n+t 2^{k}} \) positive real numbers,
    \begin{align} \pmb{ \frac{a_{1}+a_{2}+…+a_{n}+tm}{n+t}\geq \sqrt[n+t]{a_{1}a_{2}…a_{n}.m^{t}} } \end{align} Substituting (9) in (10), we get \begin{align*} &\pmb{ \frac{nm+tm}{n+t}\geq \sqrt[n+t]{a_{1}a_{2}…a_{n}.m^{t}}}\\ \implies & \pmb{ m\geq \sqrt[n+t]{a_{1}a_{2}…a_{n}.m^{t}} }\\ \implies & \pmb{ m^{n+t}\geq a_{1}a_{2}…a_{n}.m^{t} }\\ \implies & \pmb{ m^{n}m^{t}\geq a_{1}a_{2}…a_{n}.m^{t} }\\ \implies & \pmb{ m^{n}\geq a_{1}a_{2}…a_{n} }\\ \implies & \pmb{ m\geq \sqrt[n]{a_{1}a_{2}…a_{n}} }\\ \implies & \pmb{ \frac{a_{1}+a_{2}+…+a_{n}}{n}\geq \sqrt[n]{a_{1}a_{2}…a_{n}} } \end{align*} The equality occurs when \(\pmb{a_{1}=a_{2}=…=a_{n}} \).

  Therefore \( \pmb{A.M.\geq G.M.} \)

Theorem-2: \( \pmb{G.M.\geq H.M.} \)

  Statement:

  Let \(\pmb{a_{1},a_{2},…,a_{n}} \) be \(\pmb{n} \) positive real numbers then \(\pmb{ \sqrt[n]{a_{1}a_{2}…a_{n}} \geq \frac{n}{\frac{1}{a_{1}}+\frac{1}{a_{2}}+…+\frac{1}{a_{n}}} }\). The equality occurs when \(\pmb{a_{1}=a_{2}=…=a_{n}} \).


  Proof:
  Let \(\pmb{a_{1},a_{2},…,a_{n}} \) be \(\pmb{n} \) positive real numbers.
  Then \(\pmb{\frac{1}{a_{1}},\frac{1}{a_{2}},…,\frac{1}{a_{n}}} \) are also \(\pmb{n} \) positive real numbers.
   Since \( \pmb{A.M.\geq G.M.} \) then \begin{align*} &\pmb{ \frac{\frac{1}{a_{1}}+\frac{1}{a_{2}}+…+\frac{1}{a_{n}}}{n}\geq \sqrt[n]{\frac{1}{a_{1}}.\frac{1}{a_{2}}…\frac{1}{a_{n}} }}\\ \implies & \pmb{ \frac{\frac{1}{a_{1}}+\frac{1}{a_{2}}+…+\frac{1}{a_{n}}}{n}\geq \frac{1}{\sqrt[n]{a_{1}a_{2}…a_{n}}} }\\ \implies & \pmb{ \sqrt[n]{a_{1}a_{2}…a_{n}} \geq \frac{n}{\frac{1}{a_{1}}+\frac{1}{a_{2}}+…+\frac{1}{a_{n}}} } \end{align*}   The equality occurs when \(\pmb{a_{1}=a_{2}=…=a_{n}} \).
  Therefore \( \pmb{G.M.\geq H.M.} \)

Applications

  • Arithmetic Mean: Used in statistics for calculating averages and summarizing data sets.
  • Geometric Mean: Commonly applied in financial analysis, particularly in compound interest and growth rates.
  • Harmonic Mean: Essential in computing average rates, such as speed, or in physics and engineering.

Conclusion

  Arithmetic, Geometric, Harmonic Means form the cornerstone of mathematical analysis, providing powerful tools to summarize and analyze data. Mastery of these means not only enhances mathematical skills but also provides insights into real-world phenomena, improving decision-making across disciplines.

References

  1. Inequalities by G.H. Hardy
  2. Mathematical Inequalities by Dragoslav Mitrinovic
  3. The Cauchy-Schwarz Master Class by J. Michael Steele
  4. Convex Functions and Their Applications by Constantin Niculescu
  5. Inequalities in Analysis and Probability by Brannan and Hayman
  6. Advanced Mathematical Inequalities by George A. Anastassiou
  7. Problem Solving Strategies by Arthur Engel
  8. Geometric Inequalities by N. D. Kazarinoff
  9. Introduction to Mathematical Inequalities by Edwin Beckenbach
  10. Topics in Inequalities by Hojoo Lee
  11. Algebraic Inequalities by Ji Chen
  12. Basic Inequalities by V. V. Prasolov
  13. Inequalities with Applications by Z. Tomovski
  14. Numerical Inequalities by D.S. Mitrinovic
  15. Problem-Solving Techniques in Inequalities by Thomas Mildorf
  16. Convex Optimization by Stephen Boyd
  17. Schur’s Inequality by Ivan Niven
  18. Optimization and Inequalities by Stephen Barnett
  19. Applied Inequalities by Marvin Marcus
  20. Functional Analysis and Inequalities by Peter Lax

Related Articles

  • Mappings
  • Binary Compositions
  • Vector Space
  • Linear Transformations

FAQs

  1. What is a subspace?
    A subspace is a subset of a vector space that is closed under vector addition and scalar multiplication.
  2. How do you determine if a set is a subspace?
    To determine if a set is a subspace, check if it contains the zero vector and if it is closed under addition and scalar multiplication.
  3. Can a single vector be a subspace?
    Yes, any single non-zero vector along with the zero vector forms a subspace.
  4. What is the dimension of a subspace?
    The dimension of a subspace is the number of vectors in its basis.
  5. Are all subsets of a vector space subspaces?
    No, not all subsets of a vector space are subspaces.
  6. What is the relationship between subspaces and linear independence?
    Linear independence is a property that can help in determining a basis for a subspace.
  7. How are subspaces applied in computer graphics?
    Subspaces are used to model and transform 3D environments.
  8. What is the intersection of two subspaces?
    The intersection of two subspaces is also a subspace.
  9. How are subspaces used in data science?
    Subspaces are applied in techniques like PCA (Principal Component Analysis) for dimensionality reduction.
  10. Can subspaces exist in infinite-dimensional spaces?
    Yes, subspaces can exist in both finite and infinite-dimensional vector spaces.
Scroll to Top