Arithmetic function

From Wikipedia, the free encyclopedia
Jump to: navigation, search

In number theory, an arithmetic, arithmetical, or number-theoretic function[1][2] is a real or complex valued function ƒ(n) defined on the set of natural numbers (i.e. positive integers) that "expresses some arithmetical property of n."[3]

An example of an arithmetic function is the non-principal character (mod 4) defined by

\chi(n) =
\;\;\,0 & \mbox{if } n \mbox{ is even}, \\
\;\;\, 1 & \mbox{if } n \equiv 1 \mod 4, \\
 -1 & \mbox{if } n \equiv 3 \mod 4.
    where (\tfrac{-4}{n})\ is the Kronecker symbol.

To emphasize that they are being thought of as functions rather than sequences, values of an arithmetic function are usually denoted by a(n) rather than an.

There is a larger class of number-theoretic functions that do not fit the above definition, e.g. the prime-counting functions. This article provides links to functions of both classes.


\sum_p f(p)\;   and   \prod_p f(p)\;   mean that the sum or product is over all prime numbers:

\sum_p f(p) = f(2) + f(3) + f(5) + \cdots     \prod_p f(p)= f(2)f(3)f(5)\ldots.

Similarly,   \sum_{p^k} f(p)\;   and   \prod_{p^k} f(p)\;   mean that the sum or product is over all prime powers with strictly positive exponent (so 1 is not counted):

\sum_{p^k} f(p) = f(2) + f(3) + f(4) +f(5) +f(7)+f(8)+f(9)+\cdots

\sum_{d|n} f(d)\;   and   \prod_{d|n} f(d)\;   mean that the sum or product is over all positive divisors of n, including 1 and n. E.g., if n = 12,

\prod_{d|12} f(d) = f(1)f(2) f(3)  f(4)  f(6)  f(12).\

The notations can be combined:   \sum_{p|n} f(p)\;   and   \prod_{p|n} f(p)\;   mean that the sum or product is over all prime divisors of n. E.g., if n = 18,

\sum_{p|18} f(p) = f(2) + f(3),\

and similarly   \sum_{p^k|n} f(p^k)\;   and   \prod_{p^k|n} f(p^k)\;   mean that the sum or product is over all prime powers dividing n. E.g., if n = 24,

\prod_{p^k|24} f(p^k) = f(2) f(3)  f(4) f(8).\

Multiplicative and additive functions[edit]

An arithmetic function a is

Two whole numbers m and n are called coprime if their greatest common divisor is 1; i.e., if there is no prime number that divides both of them.

Then an arithmetic function a is

  • additive if a(mn) = a(m) + a(n) for all coprime natural numbers m and n;
  • multiplicative if a(mn) = a(m)a(n) for all coprime natural numbers m and n.

Ω(n), ω(n), νp(n) – prime power decomposition[edit]

The fundamental theorem of arithmetic states that any positive integer n can be represented uniquely as a product of powers of primes:    n = p_1^{a_1}\ldots p_k^{a_k}   where p1 < p2 < ... < pk are primes and the aj are positive integers. (1 is given by the empty product.)

It is often convenient to write this as an infinite product over all the primes, where all but a finite number have a zero exponent. Define νp(n) as the exponent of the highest power of the prime p that divides n. I.e. if p is one of the pi then νp(n) = ai, otherwise it is zero. Then

n=\prod_p p^{\nu_p(n)}.

In terms of the above the functions ω and Ω are defined by

ω(n) = k,
Ω(n) = a1 + a2 + ... + ak.

To avoid repetition, whenever possible formulas for the functions listed in this article are given in terms of n and the corresponding pi, ai, ω, and Ω.

The chaotic course of Ω(n) through the natural numbers (OEISA001222): Beginning on the height of the red line the two least significant binary digits of Ω(n) of all positive odd n below 1200 are represented by a line up (digit 1) or a line down (digit 0). The additional replacement of the “↗↘” of the semiprimes without prime factors below 5 by only one line down (here blue) even almost brings a balance between the ups and downs. The prime numbers are marked orange: orange lines for the Gaussian primes and for the other primes p additional the number x in orange so that p = x² + y² = (x + yi) (x – yi) = –i (x + yi) (y + xi) and x < y for natural x and y (OEISA002331).

Multiplicative functions[edit]

σk(n), τ(n), d(n) – divisor sums[edit]

σk(n) is the sum of the kth powers of the positive divisors of n, including 1 and n, where k is a complex number.

σ1(n), the sum of the (positive) divisors of n, is usually denoted by σ(n).

Since a positive number to the zero power is one, σ0(n) is therefore the number of (positive) divisors of n; it is usually denoted by d(n) or τ(n) (for the German Teiler = divisors).

\sigma_k(n) = \prod_{i=1}^{\omega(n)} \frac{p_i^{(a_i+1)k}-1}{p_i^k-1}
= \prod_{i=1}^{\omega(n)} \left(1 + p_i^k + p_i^{2k} + \cdots + p_i^{a_i k}\right).

Setting k = 0 in the second product gives

\tau(n) = d(n) = (1 + a_{1})(1+a_{2})\cdots(1+a_{\omega(n)}).

φ(n) – Euler totient function[edit]

φ(n), the Euler totient function, is the number of positive integers not greater than n that are coprime to n.

\varphi(n) = n \prod_{p|n} \left(1-\frac{1}{p}\right)
=n \left(\frac{p_1 - 1}{p_1}\right)\left(\frac{p_2 - 1}{p_2}\right) \ldots \left(\frac{p_{\omega(n)} - 1}{p_{\omega(n)}}\right)

Jk(n) – Jordan totient function[edit]

Jk(n), the Jordan totient function, is the number of k-tuples of positive integers all less than or equal to n that form a coprime (k + 1)-tuple together with n. It is a generalization of Euler's totient, φ(n) = J1(n).

J_k(n) = n^k \prod_{p|n} \left(1-\frac{1}{p^k}\right)
=n^k \left(\frac{p^k_1 - 1}{p^k_1}\right)\left(\frac{p^k_2 - 1}{p^k_2}\right) \ldots \left(\frac{p^k_{\omega(n)} - 1}{p^k_{\omega(n)}}\right)

μ(n) - Möbius function[edit]

μ(n), the Möbius function, is important because of the Möbius inversion formula. See Dirichlet convolution, below.

\mu(n)=\begin{cases} (-1)^{\omega(n)}=(-1)^{\Omega(n)} &\mbox{if }\; \omega(n) = \Omega(n)\\
0&\mbox{if }\;\omega(n) \ne \Omega(n).\end{cases}

This implies that μ(1) = 1. (Because Ω(1) = ω(1) = 0.)

τ(n) – Ramanujan tau function[edit]

τ(n), the Ramanujan tau function, is defined by its generating function identity:

\sum_{n\geq 1}\tau(n)q^n=q\prod_{n\geq 1}(1-q^n)^{24}.

Although it is hard to say exactly what "arithmetical property of n" it "expresses",[4] (τ(n) is (2π)−12 times the nth Fourier coefficient in the q-expansion of the modular discriminant function)[5] it is included among the arithmetical functions because it is multiplicative and it occurs in identities involving certain σk(n) and rk(n) functions (because these are also coefficients in the expansion of modular forms).

cq(n) – Ramanujan's sum[edit]

cq(n), Ramanujan's sum, is the sum of the nth powers of the primitive qth roots of unity:

\sum_{\stackrel{1\le a\le q}{ \gcd(a,q)=1}}
e^{2 \pi i \tfrac{a}{q} n}

Even though it is defined as a sum of complex numbers (irrational for most values of q), it is an integer. For a fixed value of n it is multiplicative in q:

If q and r are coprime, c_q(n)c_r(n)=c_{qr}(n).\;

Many of the functions mentioned in this article have expansions as series involving these sums; see the article Ramanujan's sum for examples.

Completely multiplicative functions[edit]

λ(n) – Liouville function[edit]

λ(n), the Liouville function, is defined by

\lambda (n) = (-1)^{\Omega(n)}.\;

χ(n) – characters[edit]

All Dirichlet characters χ(n) are completely multiplicative. An example is the non-principal character (mod 4) defined in the introduction. Two characters have special notations:

The principal character (mod n) is denoted by χ0(a) (or χ1(a)). It is defined as

 \chi_0(a) = \begin{cases} 1 & \mbox{if } \gcd(a,n) = 1, \\ 0 & \mbox{if } \gcd(a,n) \ne 1.

The quadratic character (mod n) is denoted by the Jacobi symbol for odd n (it is not defined for even n.):

\Bigg(\frac{a}{n}\Bigg) = \left(\frac{a}{p_1}\right)^{a_1}\left(\frac{a}{p_2}\right)^{a_2}\cdots \left(\frac{a}{p_{\omega(n)}}\right)^{a_{\omega(n)}}.

In this formula (\tfrac{a}{p}) is the Legendre symbol, defined for all integers a and all odd primes p by

\left(\frac{a}{p}\right) = \begin{cases}
\;\;\,0\mbox{ if } a \equiv 0 \pmod{p}
\\+1\mbox{ if }a \not\equiv 0\pmod{p} \mbox{ and for some integer }x, \;a\equiv x^2\pmod{p}
\\-1\mbox{ if there is no such } x. \end{cases}

Following the normal convention for the empty product, \left(\frac{a}{1}\right) = 1.

Additive functions[edit]

ω(n) – distinct prime divisors[edit]

ω(n), defined above as the number of distinct primes dividing n, is additive.

Completely additive functions[edit]

Ω(n) – prime divisors[edit]

Ω(n), defined above as the number of prime factors of n counted with multiplicities, is completely additive.

νp(n) – prime power dividing n[edit]

For a fixed prime p, νp(n), defined above as the exponent of the largest power of p dividing n, is completely additive.

Neither multiplicative nor additive[edit]

π(x), Π(x), θ(x), ψ(x) – prime count functions[edit]

Unlike the other functions listed in this article, these are defined for non-negative real (not just integer) arguments. They are used in the statement and proof of the prime number theorem.

π(x), the prime counting function, is the number of primes not exceeding x.

\pi(x) = \sum_{p\le x}1

A related function counts prime powers with weight 1 for primes, 1/2 for their squares, 1/3 for cubes, ...

\Pi(x) = \sum_{p^k\le x}\frac{1}{k}.

θ(x) and ψ(x), the Chebyshev functions are defined as sums of the natural logarithms of the primes not exceeding x:

\vartheta(x)=\sum_{p\le x} \log p,
 \psi(x) = \sum_{p^k\le x} \log p.

Λ(n) – von Mangoldt function[edit]

Λ(n), the von Mangoldt function, is 0 unless the argument is a prime power, in which case it is the natural log of the prime:

\Lambda(n) = \begin{cases}\log p &\mbox{if } n = 2,3,4,5,7,8,9,11,13,16,\ldots=p^k \mbox{ is a prime power}\\
0&\mbox{if } n=1,6,10,12,14,15,18,20,21,\dots \;\;\;\;\mbox{ is not a prime power}.

p(n) – partition function[edit]

p(n), the partition function, is the number of ways of representing n as a sum of positive integers, where two representations with the same summands in a different order are not counted as being different:

p(n) = |\left\{ (a_1, a_2,\dots a_k): 0 < a_1 \le a_2 \le \ldots \le a_k\; \and \;n=a_1+a_2+\cdots +a_k  \right\}|.

λ(n) – Carmichael function[edit]

λ(n), the Carmichael function, is the smallest positive number such that a^{\lambda(n)}\equiv 1 \pmod{n}   for all a coprime to n. Equivalently, it is the least common multiple of the orders of the elements of the multiplicative group of integers modulo n.

For powers of odd primes and for 2 and 4, λ(n) is equal to the Euler totient function of n; for powers of 2 greater than 4 it is equal to one half of the Euler totient function of n:

\lambda(n) =
\;\;\phi(n) &\mbox{if }n = 2,3,4,5,7,9,11,13,17,19,23,25,27,\dots\\
\tfrac12\phi(n)&\text{if }n=8,16,32,64,\dots

and for general n it is the least common multiple of λ of each of the prime power factors of n:

\lambda(p_1^{a_1}p_2^{a_2} \dots p_{\omega(n)}^{a_{\omega(n)}}) = \operatorname{lcm}[\lambda(p_1^{a_1}),\;\lambda(p_2^{a_2}),\dots,\lambda(p_{\omega(n)}^{a_{\omega(n)}}) ].

h(n) – Class number[edit]

h(n), the class number function, is the order of the ideal class group of an algebraic extension of the rationals with discriminant n. The notation is ambiguous, as there are in general many extensions with the same discriminant. See quadratic field and cyclotomic field for classical examples.

rk(n) – Sum of k squares[edit]

rk(n) is the number of ways n can be represented as the sum of k squares, where representations that differ only in the order of the summands or in the signs of the square roots are counted as different.

r_k(n) = |\{(a_1, a_2,\dots,a_k):n=a_1^2+a_2^2+\cdots+a_k^2\}|

Summation functions[edit]

Given an arithmetic function a(n), its summation function A(x) is defined by

 A(x) := \sum_{n \le x} a(n) .

A can be regarded as a function of a real variable. Given a positive integer m, A is constant along open intervals m < x < m + 1, and has a jump discontinuity at each integer for which a(m) ≠ 0.

Since such functions are often represented by series and integrals, to achieve pointwise convergence it is usual to define the value at the discontinuities as the average of the values to the left and right:

 A_0(m) := \frac12\left(\sum_{n < m} a(n) +\sum_{n \le m} a(n)\right) = A(m) - \frac12 a(m) .

Individual values of arithmetic functions may fluctuate wildly – as in most of the above examples. Summation functions "smooth out" these fluctuations. In some cases it may be possible to find asymptotic behaviour for the summation function for large x.

A classical example of this phenomenon[6] is given by the divisor summatory function, the summation function of d(n), the number of divisors of n:

\liminf_{n\to\infty}d(n) = 2
\limsup_{n\to\infty}\frac{\log d(n) \log\log n}{\log n} = \log 2
\lim_{n\to\infty}\frac{d(1) + d(2)+ \cdots +d(n)}{\log(1) + \log(2)+ \cdots +\log(n)} = 1.

The average order of an arithmetic function is some simpler or better-understood function which has the same summation function asmyptotically, and hence takes the same values "on average". We say that the average order of f is g if

 \sum_{n \le x} f(n) \sim \sum_{n \le x} g(n)

as x tends to infinity. The example above shows that d(n) has the average order log(n).[7]

Dirichlet convolution[edit]

Given an arithmetic function a(n), let Fa(s), for complex s, be the function defined by the corresponding Dirichlet series (where it converges):[8]

 F_a(s) := \sum_{n=1}^{\infty} \frac{a(n)}{n^s} .

Fa(s) is called a generating function of a(n). The simplest such series, corresponding to the constant function a(n) = 1 for all n, is ς(s) the Riemann zeta function.

The generating function of the Möbius function is the inverse of the zeta function:

\zeta(s)\,\sum_{n=1}^\infty\frac{\mu(n)}{n^s}=1, \;\;\mathfrak{R} \,s >0.

Consider two arithmetic functions a and b and their respective generating functions Fa(s) and Fb(s). The product Fa(s)Fb(s) can be computed as follows:

 F_a(s)F_b(s) = \left( \sum_{m=1}^{\infty}\frac{a(m)}{m^s} \right)\left( \sum_{n=1}^{\infty}\frac{b(n)}{n^s} \right) .

It is a straightforward exercise to show that if c(n) is defined by

 c(n) := \sum_{ij = n} a(i)b(j) = \sum_{i|n}a(i)b\left(\frac{n}{i}\right) ,


F_c(s) =F_a(s) F_b(s).\;

This function c is called the Dirichlet convolution of a and b, and is denoted by a*b.

A particularly important case is convolution with the constant function a(n) = 1 for all n, corresponding to multiplying the generating function by the zeta function:

g(n) = \sum_{d|n}f(d).\;

Multiplying by the inverse of the zeta function gives the Möbius inversion formula:

f(n) = \sum_{d|n}\mu\left(\frac{n}{d}\right)g(d).

If f is multiplicative, then so is g. If f is completely multiplicative, then g is multiplicative, but may or may not be completely multiplicative. The article multiplicative function has a short proof.

Relations among the functions[edit]

There are a great many formulas connecting arithmetical functions with each other and with the functions of analysis, especially powers, roots, and the exponential and log functions.

Here are a few examples:

Dirichlet convolutions[edit]

\sum_{\delta\mid n}\mu(\delta)=
\sum_{\delta\mid n}\lambda\left(\frac{n}{\delta}\right)|\mu(\delta)|=
&1\mbox{ if } n=1\\
&0\mbox{ if } n\ne1.
    where λ is the Liouville function.    [9]

\sum_{\delta\mid n}\varphi(\delta)=
=\sum_{\delta\mid n}\mu\left(\frac{n}{\delta}\right)\delta
=n\sum_{\delta\mid n}\frac{\mu(\delta)}{\delta}.
      Möbius inversion

\sum_{d | n } J_k(d) = n^k. \, 

=\sum_{\delta\mid n}\mu\left(\frac{n}{\delta}\right)\delta^k
=n^k\sum_{\delta\mid n}\frac{\mu(\delta)}{\delta^k}.
      Möbius inversion

\sum_{\delta\mid n}\delta^sJ_r(\delta)J_s\left(\frac{n}{\delta}\right) = J_{r+s}(n)

\sum_{\delta\mid n}\varphi(\delta)d\left(\frac{n}{\delta}\right)=

\sum_{\delta\mid n}|\mu(\delta)|=
|\mu(n)|=\sum_{\delta\mid n}\mu\left(\frac{n}{\delta}\right)2^{\omega(\delta)}.
      Möbius inversion

\sum_{\delta\mid n}2^{\omega(\delta)}=
2^{\omega(n)}=\sum_{\delta\mid n}\mu\left(\frac{n}{\delta}\right)d(\delta^2).
      Möbius inversion

\sum_{\delta\mid n}d(\delta^2)=
d(n^2)=\sum_{\delta\mid n}\mu\left(\frac{n}{\delta}\right)d^2(\delta).
      Möbius inversion

\sum_{\delta\mid n}d\left(\frac{n}{\delta}\right)2^{\omega(\delta)}=

\sum_{\delta\mid n}\lambda(\delta)=\begin{cases}
&1\mbox{ if } n \mbox{ is a square }\\
&0\mbox{ if } n \mbox{ is not  square.}
    where λ is the Liouville function.

\sum_{\delta\mid n}\Lambda(\delta)=
\log n.
\Lambda(n)=\sum_{\delta\mid n}\mu\left(\frac{n}{\delta}\right)\log(\delta).
      Möbius inversion

Sums of squares[edit]

\mbox{If }k \ge 4,\;\;\; r_k(n) > 0.     (Lagrange's four-square theorem).

r_2(n) = 4\sum_{d|n}\chi(d),\;     where χ is the non-principal character (mod 4) defined in the introduction.[17]

There is a formula for r3 in the section on class numbers below.

r_4(n) =
8 \sum_{\stackrel{d\,|\,n}{ 4\, \nmid \,d}}d =
8 (2+(-1)^n)\sum_{\stackrel{d\,|\,n}{ 2\, \nmid \,d}}d =
8\sigma(n)&\mbox{if } n \mbox{ is odd }\\
24\sigma\left(\frac{n}{2^{\nu}}\right)&\mbox{if } n \mbox{ is even }
    where ν = ν2(n).    [18][19][20]

r_6(n) = 16 \sum_{d|n} \chi\left(\frac{n}{d}\right)d^2 - 4\sum_{d|n} \chi(d)d^2.

Define the function σk*(n) as[22]

\sigma_k^*(n) = (-1)^{n}\sum_{d|n}(-1)^d d^k=
\sum_{d\,|\,n} d^k=\sigma_k(n)&\mbox{if } n \mbox{ is odd }\\
\sum_{\stackrel{d\,|\,n}{ 2\, \mid \,d}}d^k -\sum_{\stackrel{d\,|\,n}{ 2\, \nmid \,d}}d^k&\mbox{if } n \mbox{ is even}.

That is, if n is odd, σk*(n) is the sum of the kth powers of the divisors of n, i.e. σk(n), and if n is even it is the sum of the kth powers of the even divisors of n minus the sum of the kth powers of the odd divisors of n.

r_8(n) = 16\sigma_3^*(n).\;

Adopt the convention that Ramanujan's τ(x) = 0 if x is not an integer.

r_{24}(n) = \frac{16}{691}\sigma_{11}^*(n) + \frac{128}{691}\left\{

Divisor sum convolutions[edit]

Here "convolution" does not mean "Dirichlet convolution" but instead refers to the formula for the coefficients of the product of two power series:

  \left(\sum_{n=0}^\infty a_n x^n\right)\left(\sum_{n=0}^\infty b_n x^n\right)
= \sum_{i=0}^\infty \sum_{j=0}^\infty  a_i b_j x^{i+j}
= \sum_{n=0}^\infty \left(\sum_{i=0}^n a_i b_{n-i}\right) x^n
= \sum_{n=0}^\infty c_n x^n

The sequence c_n = \sum_{i=0}^n a_i b_{n-i}\; is called the convolution or the Cauchy product of the sequences an and bn.
See Eisenstein series for a discussion of the series and functional identities involved in these formulas.[25]

\sigma_3(n) = \frac{1}{5}\left\{6n\sigma_1(n)-\sigma_1(n) + 12\sum_{0<k<n}\sigma_1(k)\sigma_1(n-k)\right\}.\;

\sigma_5(n) = \frac{1}{21}\left\{10(3n-1)\sigma_3(n)+\sigma_1(n) + 240\sum_{0<k<n}\sigma_1(k)\sigma_3(n-k)\right\}.\;

&=\frac{1}{20}\left\{21(2n-1)\sigma_5(n)-\sigma_1(n) + 504\sum_{0<k<n}\sigma_1(k)\sigma_5(n-k)\right\}\\
&=\sigma_3(n) + 120\sum_{0<k<n}\sigma_3(k)\sigma_3(n-k).

&= \frac{1}{11}\left\{10(3n-2)\sigma_7(n)+\sigma_1(n) + 480\sum_{0<k<n}\sigma_1(k)\sigma_7(n-k)\right\}\\
&= \frac{1}{11}\left\{21\sigma_5(n)-10\sigma_3(n) + 5040\sum_{0<k<n}\sigma_3(k)\sigma_5(n-k)\right\}.\;

\tau(n) = \frac{65}{756}\sigma_{11}(n) + \frac{691}{756}\sigma_{5}(n) - \frac{691}{3}\sum_{0<k<n}\sigma_5(k)\sigma_5(n-k),\;
    where τ(n) is Ramanujan's function.    [30][31]

Since σk(n) (for natural number k) and τ(n) are integers, the above formulas can be used to prove congruences[32] for the functions. See Tau-function for some examples.

Extend the domain of the partition function by setting p(0) = 1.

p(n)=\frac{1}{n}\sum_{1\le k\le n}\sigma(k)p(n-k).
   [33]   This recurrence can be used to compute p(n).

Class number related[edit]

Peter Gustav Lejeune Dirichlet discovered formulas that relate the class number h of quadratic number fields to the Jacobi symbol.[34]

An integer D is called a fundamental discriminant if it is the discriminant of a quadratic number field. This is equivalent to D ≠ 1 and either a) D is squarefree and D ≡ 1 (mod 4) or b) D ≡ 0 (mod 4), D/4 is squarefree, and D/4 ≡ 2 or 3 (mod 4).[35]

Extend the Jacobi symbol to accept even numbers in the "denominator" by defining the Kronecker symbol:

\left(\frac{a}{2}\right) = \begin{cases}
\;\;\,0&\mbox{ if } a \mbox{ is even}
\\(-1)^{\frac{a^2-1}{8}}&\mbox{ if }a \mbox{ is odd. }

Then if D < −4 is a fundamental discriminant[36][37]

h(D) & = \frac{1}{D} \sum_{r=1}^{|D|}r\left(\frac{D}{r}\right)\\
     & = \frac{1}{2-\left(\tfrac{D}{2}\right)} \sum_{r=1}^{|D|/2}\left(\frac{D}{r}\right).

There is also a formula relating r3 and h. Again, let D be a fundamental discriminant, D < −4. Then[38]

r_3(|D|) = 12\left(1-\left(\frac{D}{2}\right)\right)h(D).

Prime-count related[edit]

Let H_n = 1 + \frac12 + \frac13 + \cdots +\frac{1}{n}   be the nth harmonic number.   Then

 \sigma(n) \le H_n + e^{H_n}\log H_n   is true for every natural number n if and only if the Riemann hypothesis is true.    [39]

The Riemann hypothesis is also equivalent to the statement that, for all n > 5040,

\sigma(n) < e^\gamma n \log \log n \,     (where γ is the Euler–Mascheroni constant).     This is Robin's theorem.

\sum_{p}\nu_p(n) = \Omega(n).\;

\psi(x)=\sum_{n\le x}\Lambda(n). \;

\Pi(x)= \sum_{n\le x}\frac{\Lambda(n)}{\log n}.\;

e^{\theta(x)}=\prod_{p\le x}p.\;

e^{\psi(x)}= \operatorname{lcm}[1,2,\dots,\lfloor x\rfloor].\;

Menon's identity[edit]

In 1965 P. Kesava Menon proved[44]

\sum_{\stackrel{1\le k\le n}{ \gcd(k,n)=1}} \gcd(k-1,n)

This has been generalized by a number of mathematicians, e.g.:

B. Sury[45]

\sum_{\stackrel{1\le k_1, k_2, \dots, k_s\le n}{ \gcd(k_1,n)=1}} \gcd(k_1-1,k_2,\dots,k_s,n)

N. Rao[46]

\sum_{\stackrel{1\le k_1, k_2, \dots, k_s\le n}{ \gcd(k_1,k_2,\dots,k_s,n)=1}} \gcd(k_1-a_1,k_2-a_2,\dots,k_s-a_s,n)^s

where a1, a2, ..., as are integers, gcd(a1, a2, ..., as, n) = 1.

L. Tóth[47]

\sum_{\stackrel{1\le k\le m}{ \gcd(k,m)=1}} \gcd(k^2-1,m_1)\gcd(k^2-1,m_2)
=\varphi(n)\sum_{\stackrel{d_1\mid m_1} {d_2\mid m_2}} \varphi(\gcd(d_1, d_2))2^{\omega(\operatorname{lcm}(d_1, d_2))},

where m1 and m2 are odd, m = lcm(m1, m2).

In fact, if f is any arithmetical function[48][49]

\sum_{\stackrel{1\le k\le n}{ \gcd(k,n)=1}} f(\gcd(k-1,n))
=\varphi(n)\sum_{d\mid n}\frac{(\mu*f)(d)}{\varphi(d)},

where * stands for Dirichlet convolution.


Let m and n be distinct, odd, and positive. Then the Jacobi symbol satisfies the Law of Quadratic Reciprocity:

 \left(\frac{m}{n}\right) \left(\frac{n}{m}\right) = (-1)^{(m-1)(n-1)/4}.    
\begin{align}&|\lambda(n)|\;\mu(n) \\&=\lambda(n)\;|\mu(n)|\\& = \mu(n), \end{align}
    and     \begin{align}&\lambda(n)\mu(n) \\&= |\mu(n)| \\&=\mu^2(n).\end{align}
    where λ(n) is Liouville's function.

2^{\omega(n)}\le d(n)\le2^{\Omega(n)}.\;
\lambda(n)\mid \phi(n)     where λ(n) is Carmichael's function. Further,
\lambda(n)= \phi(n) \mbox{ if and only if }n=\begin{cases}1,2, 4\\
3,5,7,9,11, \ldots \mbox{ i.e. } p^k \mbox{ where }p\mbox{ is an odd prime}\\
6,10,14,18,\ldots \mbox{ i.e. } 2p^k\mbox{ where }p\mbox{ is an odd prime}


&=\frac{\mu\left(\frac{q}{\gcd(q, n)}\right)}{\phi\left(\frac{q}{\gcd(q, n)}\right)}\phi(q)\\
   [54]     Note that  \phi(q) = \sum_{\delta|q}\mu\left(\frac{q}{\delta}\right)\delta.    [55]
c_q(1) = \mu(q).\;
c_q(q) = \phi(q).\;

\sum_{\delta|n}d^{\;3}(\delta) = \left(\sum_{\delta|n}d(\delta)\right)^2.\;
   [56]   Compare this with 13 + 23 + 33 + ... + n3 = (1 + 2 + 3 + ... + n)2
d(uv) = \sum_{\delta\;|\gcd(u,v)}\mu(\delta)d\left(\frac{u}{\delta}\right)d\left(\frac{v}{\delta}\right).\;
\sigma_k(u)\sigma_k(v) = \sum_{\delta\;|\gcd(u,v)}\delta^k\sigma_k\left(\frac{uv}{\delta^2}\right).\;
\tau(u)\tau(v) = \sum_{\delta\;|\gcd(u,v)}\delta^{11}\tau\left(\frac{uv}{\delta^2}\right),\;
    where τ(n) is Ramanujan's function.    [59]


  1. ^ Long (1972, p. 151)
  2. ^ Pettofrezzo & Byrkit (1970, p. 58)
  3. ^ Hardy & Wright, intro. to Ch. XVI
  4. ^ Hardy, Ramanujan, § 10.2
  5. ^ Apostol, Modular Functions ..., § 1.15, Ch. 4, and ch. 6
  6. ^ Hardy & Wright, §§ 18.1–18.2
  7. ^ Gérald Tenenbaum (1995). Introduction to Analytic and Probabilistic Number Theory. Cambridge studies in advanced mathematics 46. Cambridge University Press. pp. 36–55. ISBN 0-521-41261-7. 
  8. ^ Hardy & Wright, § 17.6, show how the theory of generating functions can be constructed in a purely formal manner with no attention paid to convergence.
  9. ^ Hardy & Wright, Thm. 263
  10. ^ Hardy & Wright, Thm. 63
  11. ^ see references at Jordan's totient function
  12. ^ Holden et al in external links The formula is Gegenbauer's
  13. ^ Hardy & Wright, Thm. 288–290
  14. ^ Dineva in external links, prop. 4
  15. ^ Hardy & Wright, Thm. 264
  16. ^ Hardy & Wright, Thm. 296
  17. ^ Hardy & Wright, Thm. 278
  18. ^ Hardy & Wright, Thm. 386
  19. ^ Hardy, Ramanujan, eqs 9.1.2, 9.1.3
  20. ^ Koblitz, Ex. III.5.2
  21. ^ a b Hardy & Wright, § 20.13
  22. ^ Hardy, Ramanujan, § 9.7
  23. ^ Hardy, Ramanujan, § 9.13
  24. ^ Hardy, Ramanujan, § 9.17
  25. ^ The paper by Huard, Ou, Spearman, and Williams in the external links also has proofs.
  26. ^ a b Ramanujan, On Certain Arithmetical Functions, Table IV; Papers, p. 146
  27. ^ a b Koblitz, ex. III.2.8
  28. ^ Koblitz, ex. III.2.3
  29. ^ Koblitz, ex. III.2.2
  30. ^ Koblitz, ex. III.2.4
  31. ^ Apostol, Modular Functions ..., Ex. 6.10
  32. ^ Apostol, Modular Functions..., Ch. 6 Ex. 10
  33. ^ G.H. Hardy, S. Ramannujan, Asymptotic Formulæ in Combinatory Analysis, § 1.3; in Ramannujan, Papers p. 279
  34. ^ Landau, p. 168, credits Gauss as well as Dirichlet
  35. ^ Cohen, Def. 5.1.2
  36. ^ Cohen, Corr. 5.3.13
  37. ^ see Edwards, § 9.5 exercises for more complicated formulas.
  38. ^ Cohen, Prop 5.10.3
  39. ^ See Divisor function.
  40. ^ Hardy & Wright, eq. 22.1.2
  41. ^ See prime counting functions.
  42. ^ Hardy & Wright, eq. 22.1.1
  43. ^ Hardy & Wright, eq. 22.1.3
  44. ^ László Tóth, Menon's Identity and Arithmetical Sums ..., #External links, eq. 1
  45. ^ Tóth, eq. 5
  46. ^ Tóth, eq. 3
  47. ^ Tóth, eq. 35
  48. ^ Tóth, eq. 2
  49. ^ Tóth states that Menon proved this for multiplicative f in 1965 and V. Sita Ramaiah for general f.
  50. ^ Hardy Ramanujan, eq. 3.10.3
  51. ^ Hardy & Wright, § 22.13
  52. ^ See Multiplicative group of integers modulo n and Primitive root modulo n.
  53. ^ Hardy & Wright, Thm. 329
  54. ^ Hardy & Wright, Thms. 271, 272
  55. ^ Hardy & Wright, eq. 16.3.1
  56. ^ Ramanujan, Some Formulæ in the Analytic Theory of Numbers, eq. (C); Papers p.133. A footnote says that Hardy told Ramanujan it also appears in an 1857 paper by Liouville.
  57. ^ Ramanujan, Some Formulæ in the Analytic Theory of Numbers, eq. (F); Papers p.134
  58. ^ Apostol, Modular Functions ..., ch. 6 eq. 4
  59. ^ Apostol, Modular Functions ..., ch. 6 eq. 3


  • Apostol, Tom M. (1989), Modular Functions and Dirichlet Series in Number Theory (2nd Edition), New York: Springer, ISBN 0-387-97127-0 
  • G. J. O. Jameson (2003), The Prime Number Theorem, Cambridge University Press, ISBN 0-521-89110-8 
  • Koblitz, Neal (1984), Introduction to Elliptic Curves and Modular Forms, New York: Springer, ISBN 0-387-97966-2 
  • Landau, Edmund (1966), Elementary Number Theory, New York: Chelsea 
  • Elliott Mendelson (1987), Introduction to Mathematical Logic, CRC Press, ISBN 0-412-80830-7 
  • Pettofrezzo, Anthony J.; Byrkit, Donald R. (1970), Elements of Number Theory, Englewood Cliffs: Prentice Hall, LCCN 77-81766 Check |lccn= value (help) 

External links[edit]