Euler-Macheroni constant

Real Analysis
Post Reply
User avatar
Grigorios Kostakos
Founder
Founder
Posts: 461
Joined: Mon Nov 09, 2015 1:36 am
Location: Ioannina, Greece

Euler-Macheroni constant

#1

Post by Grigorios Kostakos »

Prove that the sequence \[\gamma_{n}=\displaystyle1+\frac{1}{2}+\frac{1}{3}+\ldots+\frac{1}{n}-\log{n}\,, \quad n\in\mathbb{N}\,,\] converges.
Grigorios Kostakos
Papapetros Vaggelis
Community Team
Posts: 426
Joined: Mon Nov 09, 2015 1:52 pm

Re: Euler-Macheroni constant

#2

Post by Papapetros Vaggelis »

For all \(\displaystyle{n\in\mathbb{N}}\) , \(\displaystyle{\gamma_{n}=\sum_{k=1}^n \frac{1}{k}-\log n}\) , so

\(\displaystyle{\begin{aligned} \gamma_{n+1}-\gamma_{n}&=\sum_{k=1}^{n+1}\frac{1}{k}-\log \left(n+1\right)-\sum_{k=1}^n \frac{1}{k}+\log n\\&=\sum_{k=1}^n \frac{1}{k}+\frac{1}{n+1}-\left[\log \left(n+1\right)-\log n\right]-\sum_{k=1}^n \frac{1}{n}\\&=\frac{1}{n+1}-\log \left(\frac{n+1}{n}\right)\end{aligned}}\)

Let \(\displaystyle{f:\left(0,+\infty\right)\to \mathbb{R}\,\,,f(x)=\log x}\).

The function \(\displaystyle{x\mapsto f^\prime(x)=\frac{1}{x}\,\,,x>0}\) is strictly decreasing at \(\displaystyle{\left(0,+\infty\right)}\) .

If \(\displaystyle{n\in\mathbb{N}}\), by applying the mean value theorem for \(\displaystyle{f}\), at the interval \(\displaystyle{\left[n,n+1\right]}\), we have that

there is \(\displaystyle{t\in\left(n,n+1\right)}\) such that

\(\displaystyle{f^\prime(t)=\frac{1}{t}=\frac{\log \left(n+1\right)-\log n}{\left(n+1\right)-n}=\log \left(\frac{n+1}{n}\right)}\) , so

\(\displaystyle{\frac{1}{n+1}<t<\frac{1}{n}\Rightarrow \frac{1}{n+1}<\log \left(\frac{n+1}{n}\right)<\frac{1}{n}}\)

Therefore, \(\displaystyle{\gamma_{n+1}-\gamma_{n}=\frac{1}{n+1}-\log \left(\frac{n+1}{n}\right)<0\ \forall n\in\mathbb{N}}\) , consequently,

the sequence \(\displaystyle{\left(\gamma_{n}\right)_{n\in\mathbb{N}}}\) is strictly decreasing.

Also, for all \(\displaystyle{n\in\mathbb{N}}\) ,

\(\displaystyle{\begin{aligned} \gamma_{n}=\sum_{k=1}^n \frac{1}{k}-\log n&\geq \sum_{k=1}^n \log \left(\frac{k+1}{k}\right)-\log n\\&=\log 2+\log \frac{3}{2}+\log \frac{4}{3}+...+\log \frac{n+1}{n}-\log n \\&=\log \left(2\cdot \frac{3}{2}\cdot \frac{4}{3}\cdot \cdot \cdot \frac{n+1}{n}\right)-\log n\\&=\log \left(n+1\right)-\log n\\&=\log \left(1+\frac{1}{n}\right)\\&>\log 1=0\end{aligned}}\) .

We proved that the sequence \(\displaystyle{\left(\gamma_{n}\right)_{n\in\mathbb{N}}}\) is bounded.

Thus, the sequence \(\displaystyle{\left(\gamma_{n}\right)_{n\in\mathbb{N}}}\) converges.
User avatar
Grigorios Kostakos
Founder
Founder
Posts: 461
Joined: Mon Nov 09, 2015 1:36 am
Location: Ioannina, Greece

Re: Euler-Macheroni constant

#3

Post by Grigorios Kostakos »

Another solution:

For every \(n\in\mathbb{N}\) holds:
\begin{align*}
& \Bigl({1+\frac{1}{n}}\Bigr)^{n}<e<\Bigl({1+\frac{1}{n}}\Bigr)^{n+1}\quad\Rightarrow\quad\log\Bigl({\bigr({1+\tfrac{1}{n}}\bigr)^{n}}\Bigr)<\log{e}<\log\Bigl({\bigr({1+\tfrac{1}{n}}\bigr)^{n+1}}\Bigr)\quad\Rightarrow\\
& 0<n\log\tfrac{n+1}{n}<1<({n+1})\log\tfrac{n+1}{n}\quad\Rightarrow\quad\frac{1}{n\log\tfrac{n+1}{n}}>1>\frac{1}{({n+1})\log\tfrac{n+1}{n}}\quad\Rightarrow
\end{align*}\[\displaystyle\frac{1}{n+1}<\log\tfrac{n+1}{n}=\log({n+1})-\log{n}<\frac{1}{n}\quad(1)\,.\]
From \((1)\) we have\begin{align*}
& \log2-\log1<\frac{1}{1}=1\\
& \log3-\log2<\frac{1}{2}\\
& \dots\dots\dots\dots\dots\dots\dots\dots\dots\\
& \log n-\log({n-1})<\frac{1}{n-1}\\
& \log({n+1})-\log n<\frac{1}{n}
\end{align*}Adding the above equations, we have\begin{align*}
& \log({n+1})<\displaystyle1+\frac{1}{2}+\frac{1}{3}+\ldots+\frac{1}{n}\quad\Rightarrow\\
& \log({n+1})-\log n<\displaystyle1+\frac{1}{2}+\frac{1}{3}+\ldots+\frac{1}{n}-\log n=\gamma_{n}\,, \ n\in\mathbb{N}\,.
\end{align*}But for every \(n\in\mathbb{N}\), we have that \[\log({n+1})-\log n=\log\bigr({1+\tfrac{1}{n}}\bigr)>0\] and the sequence \(\left({\gamma_{n}}\right)_{n\in\mathbb{N}}\) is strictly decreasing because \begin{alignat*}{2}
\gamma_{n+1}-\gamma_{n} & =\displaystyle1+\frac{1}{2}+\frac{1}{3}+\ldots+\frac{1}{n}+\frac{1}{n+1}-\log({n+1})-1-\frac{1}{2}-\frac{1}{3}-\ldots-\frac{1}{n}+\log n\\
& =\frac{1}{n+1}-\log\tfrac{n+1}{n}\stackrel{(1)}{<}0\,.
\end{alignat*} So, the sequence \(\left({\gamma_{n}}\right)_{n\in\mathbb{N}}\) converges to a real number \(\gamma\), named the Euler-Macheroni constant.
Grigorios Kostakos
User avatar
Tolaso J Kos
Administrator
Administrator
Posts: 867
Joined: Sat Nov 07, 2015 6:12 pm
Location: Larisa
Contact:

Re: Euler-Macheroni constant

#4

Post by Tolaso J Kos »

Splitting hairs around here, well, here is another approach that is not that much different than the previous two answers.

To prove that the sequence converges we have to show that it is bounded and monotonic.Here we go.

Monotony: It is easy to prove that the sequence is decreasing. Indeed:

$$\gamma_{n+1} - \gamma_n = \left ( \mathcal{H}_{n+1}- \ln (n+1) \right ) - \left ( \mathcal{H}_n - \ln n \right ) = \frac{1}{n}+ \ln \left ( 1- \frac{1}{n} \right )<0 $$

The inequality is valid since $\ln (1-x)$ is a concave function hence lies beneath the line $y=-x$ that is its tangent to its graph at $x_0=0$. Plugging $x=1/n$ yields the result.

Boundness: Using the fact that:

$$\mathcal{H}_n = \sum_{k=1}^{n}\frac{1}{k}> \int_{1}^{n}\frac{{\rm d}t}{t}= \ln (n+1)> \ln n$$

we have that $\gamma_n = \mathcal{H}_n -\ln n >0$. Hence $\gamma_n$ is bounded from below.


Therefore the sequence converges to the famous Euler - Mascheroni constant. More information can be found here
Imagination is much more important than knowledge.
Post Reply

Create an account or sign in to join the discussion

You need to be a member in order to post a reply

Create an account

Not a member? register to join our community
Members can start their own topics & subscribe to topics
It’s free and only takes a minute

Register

Sign in

Who is online

Users browsing this forum: No registered users and 47 guests