Arithmetic–geometric mean

In mathematics, the arithmetic–geometric mean (AGM or agM) is a special way to find an average of two positive real numbers. It uses two sequences—one made from arithmetic means and the other from geometric means—that get closer and closer until they reach the same number. This number is called the arithmetic–geometric mean of the two starting numbers.

The AGM is very useful in creating fast algorithms for calculating things like exponential and trigonometric functions, as well as important mathematical constants, especially when trying to find a very accurate value for computing π.

To find the AGM of two numbers x and y, we start with a₀ = x and g₀ = y. Then we keep updating the numbers using the rules aₙ₊₁ = ½ (aₙ + gₙ) and gₙ₊₁ = √(aₙgₙ). As we repeat these steps, both sequences converge to the same value, which is the arithmetic–geometric mean of x and y. This idea can also be used with complex numbers in more advanced math.

Example

To find the arithmetic–geometric mean of a₀ = 24 and g₀ = 6, we follow these steps:

• The first arithmetic mean is half of (24 + 6), which equals 15.
• The first geometric mean is the square root of (24 × 6), which equals 12.
• The second arithmetic mean is half of (15 + 12), which equals 13.5.
• The second geometric mean is the square root of (15 × 12), which is about 13.416.

Each step makes the two numbers closer together. After many steps, they reach a single number, called the arithmetic–geometric mean. For 24 and 6, this number is about 13.458.

History

The first method using this special pair of sequences was created by Joseph-Louis Lagrange. Later, Carl Friedrich Gauss studied its properties and learned more about how it works.

Properties

The arithmetic mean and geometric mean of two positive numbers are always between those numbers. The geometric mean is always less than or equal to the arithmetic mean. When we repeatedly calculate these means, the geometric means increase and the arithmetic means decrease, getting closer to each other.

This process creates a special number called the arithmetic–geometric mean, which lies between the original two numbers and also between their means. It has many useful properties and can help compute important math functions and constants. It is especially good at solving problems involving elliptic integrals, which are used in designing certain filters.

Related concepts

The arithmetic–geometric mean connects to many important numbers and ideas in math. For example, the reciprocal of the arithmetic–geometric mean of 1 and the square root of 2 is called Gauss's constant. In 1799, a mathematician named Gauss showed a special link between this mean and pi, one of the most famous numbers in math.

This mean can also help us understand and calculate many useful things, like logarithms and special math functions called complete and incomplete elliptic integrals of the first and second kind and Jacobi elliptic functions.

Proof of existence

The inequality of arithmetic and geometric means shows that one sequence stays the same or gets bigger, and it never goes past a certain point. Because of this, we know the sequence will settle down to a single value. We can also see that the other sequence will end up at the same value. This proves that these special averages do exist.

Proof of the integral-form expression

This proof was created by Gauss. It shows a special way to find the arithmetic–geometric mean of two numbers using a special kind of math problem called an integral.

The proof starts with a math expression that looks at two numbers, x and y, and uses them in a special kind of sum. By changing how we look at the problem, we can see that this expression connects to the arithmetic and geometric means of x and y.

In the end, the proof shows that the arithmetic–geometric mean of x and y can be found using a simple formula that involves π (pi) and this special integral.

Applications

The number π

The number π can be found using a special way called the Gauss–Legendre algorithm. This method helps us calculate π very accurately by using a special kind of average of two numbers.

Complete elliptic integral K(sin α)

Another important use of this special average is in calculating something called the complete elliptic integral. This helps solve certain kinds of math problems that appear in many areas, like physics.

Other applications

Because of its useful properties, the special average can also help us quickly calculate important math functions, such as exponentials and trigonometric functions like sine and cosine. Many mathematicians have studied and used these methods to make calculations faster and more precise.