Sign up ×
MathOverflow is a question and answer site for professional mathematicians. It's 100% free, no registration required.

In the Princeton Companion of Mathematics, the Analytic Number Theory section, the author mentions what he calls Gauss-Kramer model, which is simply modeling the integers on a countable sequence of random variables $\{ X_{n} \}$, where the variable $X_{n}$ has probability of $1/log(n)$ of being 1 and otherwise 0. Each variable $X_{n}$ represents whether the number n is prime or not. The Gauss-Kramer model conjectures that statements about the distribution of primes correspond to the statements on the countable collection of random variables with probability 1 of being true.

Question: Is anyone aware of a good exposition of the Gauss-Kramer model and any attempt to make it more rigorous using Model Theory or through other routes?

share|cite|improve this question
Usually written Cramer. I remember this MO question $$ $$… $$ $$ I will look for others. To answer your second question, it is known to be faulty by 1985 work of Meier. – Will Jagy Jun 7 '11 at 3:24
Alright, Maier. see $$ $$… $$ $$ – Will Jagy Jun 7 '11 at 3:25
Thanks a lot for the suggestion, I am reading the synopsis of Meier's work currently. – Mohamed Alaa El Behairy Jun 7 '11 at 16:28

2 Answers 2

up vote 1 down vote accepted

Chapter 3 of The Prime Numbers and Their Distribution by Gérald Tenenbaum and Michel Mendès France has a nice exposition of the model, including modifications indicated by Maier's discovery.

share|cite|improve this answer
Thanks a lot for your suggestion. Has there been any attempt to make the model more rigorous? – Mohamed Alaa El Behairy Jun 7 '11 at 16:29
I don't think there has been much hope of making it rigorous, because the conclusions it yields are so far beyond what can be proved today. For example, if $p_k$ denotes the k-th prime, it is known that $p_{k+1} = p_k + O(p_k^{\alpha + \varepsilon})$ with $\alpha = 0.525$. The Riemann Hypothesis would allow $\alpha = 0.5$. But the Cramér model would yield $p_{k+1} = p_k + O(\log^2(p_k))$. – Marius Overholt Jun 7 '11 at 16:48

You might want to look at:

1) "Harold Cramér and the distribution of prime numbers" by Andrew Granville

2) "The distribution of prime numbers" by K. Soundararajan

share|cite|improve this answer

Your Answer


By posting your answer, you agree to the privacy policy and terms of service.

Not the answer you're looking for? Browse other questions tagged or ask your own question.