Students usually learn the following basic table of cosine function values for special points of the circle:
For real values of argument , the values of are real.
In the points , the values of are algebraic. In several cases they can even be rational numbers, 0, or 1. Here are some examples:
The values of can be expressed using only square roots if and is a product of a power of 2 and distinct Fermat primes {3, 5, 17, 257, …}.
The function is an entire analytical function of that is defined over the whole complex ‐plane and does not have branch cuts and branch points. It has an essential singular point at . It is a periodic function with the real period :
The function is an even function with mirror symmetry:
The derivatives of have simple representations using either the function or the function:
The function satisfies the simplest possible linear differential equation with constant coefficients:
The complete solution of this equation can be represented as a linear combination of and with arbitrary constant coefficients and :
The function also satisfies firstorder nonlinear differential equations:
The function has a simple series expansion at the origin that converges in the whole complex ‐plane:
For real this series can be interpreted as the real part of the series expansion for the exponential function :
The following famous infinite product representation for clearly illustrates that at :
Indefinite integrals of expressions involving the cosine function can sometimes be expressed using elementary functions. However, special functions are frequently needed to express the results even when the integrands have a simple form (if they can be evaluated in closed form). Here are some examples:
The last integral cannot be evaluated in closed form using the known classical special functions for arbitrary values of parameters and .
Definite integrals that contain the cosine function are sometimes simple. For example, the famous Dirichlet type and Fresnel integrals have the following values:
where is the Euler‐Mascheroni constant .
Some special functions can be used to evaluate more complicated definite integrals. For example, elliptic integrals and gamma functions are needed to express the following integrals:
Integral transforms of expressions involving the cosine function may not be classically convergent but can be interpreted in a generalized functions (distributions) sense. For example, the exponential Fourier transform of the cosine function does not exist in the classical sense but can be expressed using the Dirac delta function.
Among other integral transforms of the cosine function, the best known are the Fourier cosine and sine transforms, and the Laplace, Mellin, Hilbert, and Hankel transforms:
The following finite sums from the cosine can be expressed using the trigonometric functions:
The following infinite sums can be expressed using elementary functions:
The following finite products from the cosine can be expressed using trigonometric functions:
The following infinite product that contains the cosine function can be expressed using the sine function:
The cosine of a sum can be represented by the rule: "the cosine of a sum is equal to the product of the cosines minus the product of the sines." A similar rule is valid for the cosine of the difference:
In the case of multiple arguments , , , …, the function can be represented as the finite sum of terms that include powers of the sine and the cosine:
The function can also be represented as the finite sum including only the cosine of :
The cosine of the half‐angle can be represented by the following simple formula that is valid in some vertical strips:
To make this formula correct for all complex , a complicated prefactor is needed:
where contains the unit step, real part, imaginary part, and the floor functions.
The sum of two cosine functions can be described by the rule: "the sum of the cosines is equal to two times the cosine of the half‐difference multiplied by the cosine of the half‐sum." A similar rule is valid for the difference of two cosines:
The product of two cosine functions and the product of the cosine and sine have the following representations:
The integer powers of the cosine functions can be expanded as finite sums of cosine functions with multiple arguments. These sums include binomial coefficients:
The bestknown inequalities for cosine functions are the following:
There are simple relations between the function and its inverse function :
The second formula is valid at least in the vertical strip . Outside of this strip, a much more complicated relation (that contains the unit step, real part, and the floor functions) holds:
Cosine and sine functions are connected by a very simple formula including the linear function in the argument:
Another famous formula, connecting and , is shown in the well‐known Pythagorean theorem:
The last restriction on can be removed, but the formula will get a complicated coefficient with , that contains the unit step, real part, imaginary part, and the floor function:
The cosine function can also be represented using other trigonometric functions by the following formulas:
The cosine function has representations using the hyperbolic functions:
The cosine function is used throughout mathematics, the exact sciences, and engineering.
