In number theory, the Turán sieve is a technique for estimating the size of "sifted sets" of positive integers which satisfy a set of conditions which are expressed by congruences. It was developed by Pál Turán in 1934.
In terms of sieve theory the Turán sieve is of combinatorial type: deriving from a rudimentary form of the inclusion–exclusion principle. The result gives an upper bound for the size of the sifted set.
Let A be a set of positive integers ≤ x and let P be a set of primes. For each p in P, let Ap denote the set of elements of A divisible by p and extend this to let Ad be the intersection of the Ap for p dividing d, when d is a product of distinct primes from P. Further let A1 denote A itself. Let z be a positive real number and P(z) denote the product of the primes in P which are ≤ z. The object of the sieve is to estimate
S(A,P,z)=\left\vertA\setminuscuppAp\right\vert.
We assume that |Ad| may be estimated, when d is a prime p by
\left\vertAp\right\vert=
1 | |
f(p) |
X+Rp
and when d is a product of two distinct primes d = p q by
\left\vertApq\right\vert=
1 | |
f(p)f(q) |
X+Rp,q
where X = |A| and f is a function with the property that 0 ≤ f(d) ≤ 1. Put
U(z)=\sumpf(p).
Then
S(A,P,z)\le
X | |
U(z) |
+
2 | |
U(z) |
\sump\left\vertRp\right\vert+
1 | |
U(z)2 |
\sump,q\left\vertRp,q\right\vert.
. Christopher Hooley . Christopher Hooley . Applications of sieve methods to the theory of numbers . Cambridge University Press . 1976 . 0-521-20915-3. 21.