Mathematics Stack Exchange is a question and answer site for people studying math at any level and professionals in related fields. Join them; it only takes a minute:

Sign up
Here's how it works:
  1. Anybody can ask a question
  2. Anybody can answer
  3. The best answers are voted up and rise to the top

If we have that: $$\theta(x)=\sum_{p\leq x}\log p,$$ and $$\psi(x)=\sum_{n\leq x}\Lambda(n)$$ Where $\Lambda(n)=\log p $ if $n=p^m$ and $\Lambda(n)=0$ in another case.

How can I prove that :

1) $\theta(x)=\psi(x)+O(\sqrt{x})$

2) $\pi(x)=\frac{\psi(x)}{\log x}+O(\frac{x}{\log^2x})$

Being $\pi(x)=\sum_{p\leq x}1$ the function that counts the number of primes less than $x$

share|cite|improve this question
The traditional notation used for these functions is a lower case $\theta$, and $\psi$ for Chebyshev's functions, with a lower case $\pi$ for the prime counting function, the capital $\Pi$ is typically used when referring to Reimann's prime counting function. – Ethan Jun 4 '13 at 6:58
up vote 9 down vote accepted

$$\binom{2n}{n}=\frac{(2n)!}{n!^2}, \text{ is always an integer}$$ $$\text{no integer between } 1 \text{ and } n \text{ is divisible by a prime between n} \text{ and } 2n $$

$$\implies\frac{(2n)!}{n!^2}=\binom{2n}{n} \text{ is divisible by all the primes between $n$ and $2n$}$$

$$\implies\binom{2n}{n}\ge\prod_{n<p\leq 2n}p$$ $$\prod_{n<p\leq 2n}p\leq\binom{2n}{n}\leq 4^n$$ Sense $$4^n=(1+1)^{2n}=\sum_{k=0}^{2n}\binom{2n}{k}\ge\binom{2n}{n}$$ So that, $$\prod_{n<p\leq 2n}p\leq 4^n$$ $$\implies \prod_{\frac{n}{2}<p\leq n}p\leq 2^n$$ Taking logarithms gives

$$\theta(n)-\theta(\frac{n}{2})\leq \ln(2)n$$ $$\implies \theta(n)=(\theta(n)-\theta(\frac{n}{2}))+( \theta(\frac{n}{2})-\theta(\frac{n}{4}) )...\leq (1+\frac{1}{2}+\frac{1}{4}...)\ln(2)n$$ $$\implies \theta(n)\leq 2\ln(2)n=n\ln(4)$$ So that we have, $$\theta(x)\leq x\ln(4)$$ Which means $$\theta(x)=O(x)$$ Also note that $$\psi(x)=\sum_{p\leq x}\ln(p)+\sum_{p^2\leq x}\ln(p)+\sum_{p^3\leq x}\ln(p)...$$

$$=\theta(x)+\theta(x^\frac{1}{2})+\theta(x^\frac{1}{3})+...$$ $$=\theta(x)+O(\theta(x^{1/2}))=\theta(x)+O(x^{1/2})$$

So that we get $$\psi(x)=\theta(x)+O(\sqrt{x})$$

And sense $$\pi(x)+O(\sqrt{x})=\sum_{1<n\leq x}\frac{\Lambda(n)}{\ln(n)}=\frac{\psi(x)}{\ln(x)}+\int_{2}^x\frac{\psi(t)}{t\ln(t)^2} dt=\frac{\psi(x)}{\ln(x)}+O(\frac{x}{\ln(x)^2})$$

We have,


share|cite|improve this answer
Why do you have $\pi(x)+O(\sqrt{x})=\sum_{1<n\leq x}\frac{\Lambda(n)}{\log n}$? – Elmo goya Jun 4 '13 at 19:01
@Elmogoya Because $$\sum_{1<n\leq x}\frac{\Lambda(n)}{\ln(n)}=\pi(x)+\frac{\pi(x^{1/2})}{2}+\frac{\pi(x^{1/3})}{3}‌​...=\pi(x)+O(\sqrt{x})$$ – Ethan Jun 4 '13 at 21:18
And... Why do you have that $$\int_{2}^{x}\frac{\psi(t)}{t\log(t)^2}dt=O\left(\frac{x}{\log(x)^2}\right)$$ – Elmo goya Jun 4 '13 at 21:41
@Elmogoya Because $$\frac{\psi(t)}{t}=O(1)$$ and $$\int_{2}^x\frac{1}{\ln(t)^2} dt =O(\frac{x}{\ln(x)^2})$$ – Ethan Jun 4 '13 at 21:50

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.