Proving $2^{\frac{1}{2}+\frac{1}{3}+\dots+\frac{1}{n}}<n$ for All $n\ge 2$

  • MHB
  • Thread starter anemone
  • Start date
In summary, the conversation discusses the proof that $2^{\left(\dfrac{1}{2}+\dfrac{1}{3}+\dfrac{1}{4}+\cdots+\dfrac{1}{n}\right)}_{\phantom{i}}<n$ for all integer $n\ge 2$, using the fact that $\displaystyle \lim_{n \rightarrow \infty} \sum_{k=1}^{n} \frac{1}{k} - \ln n = \gamma$ where $\gamma$ is Euler's constant. It is shown that for $n$ "large enough", the logarithm of the first expression is less than the logarithm of the second expression
  • #1
anemone
Gold Member
MHB
POTW Director
3,883
115
Prove that $2^{\left(\dfrac{1}{2}+\dfrac{1}{3}+\dfrac{1}{4}+\cdots+\dfrac{1}{n}\right)}_{\phantom{i}}<n$ for all integer $n\ge 2$.
 
Mathematics news on Phys.org
  • #2
anemone said:
Prove that $2^{\left(\dfrac{1}{2}+\dfrac{1}{3}+\dfrac{1}{4}+\cdots+\dfrac{1}{n}\right)}_{\phantom{i}}<n$ for all integer $n\ge 2$.

[sp]Is...

$\displaystyle \lim_{n \rightarrow \infty} \sum_{k=1}^{n} \frac{1}{k} - \ln n = \gamma\ (1)$

... where $\gamma = .5772... $ is thye Euler's constant, so that...

$\displaystyle \lim_{n \rightarrow \infty} \sum_{k=2}^{n} \frac{1}{k} - \ln n = \gamma - 1 < 0\ (2)$

... and that means that for n 'large enough' ...

$\displaystyle \ln \{ 2^{(\frac{1}{2} + \frac{1}{3} + ... + \frac{1}{n})} \} < \ln \{ e^{(\frac{1}{2} + \frac{1}{3} + ... + \frac{1}{n})} \} < \ln n\ (3)$ [/sp]

Kind regards

$\chi$ $\sigma$
 
  • #3
we have
$(\dfrac{k}{k-1})^k = (1 + \dfrac{1}{k-1})^k \ge 1 + k \dfrac{1}{k-1} \gt 2$
or $2^\dfrac{1}{k} < (\dfrac{k}{k-1})$
multiply n- 1 terms taking k from 2 to n we get the result
 
  • #4
My solution:

We see that for the base case $n=2$, we have:

\(\displaystyle 2^{\frac{1}{2}}<2\)

This is true, so we may state the induction hypothesis $P_k$:

\(\displaystyle 2^{\sum\limits_{j=2}^k\left(\dfrac{1}{j}\right)}<k\)

If, as our induction step, we multiply $P_k$ by \(\displaystyle 2^{\dfrac{1}{k+1}}\), there results:

\(\displaystyle 2^{\sum\limits_{j=2}^{k+1}\left(\dfrac{1}{j}\right)}<k\cdot2^{\dfrac{1}{k+1}}\)

Now, consider that:

\(\displaystyle 0<\sum_{k=2}^{n+1}\left({n+1 \choose k}\frac{1}{n^k}\right)\)

Hence:

\(\displaystyle 2<2+\frac{1}{n}+\sum_{k=2}^{n+1}\left({n+1 \choose k}\frac{1}{n^k}\right)=\sum_{k=0}^{n+1}\left({n+1 \choose k}\frac{1}{n^k}\right)=\left(1+\frac{1}{n}\right)^{n+1}\)

Thus, we must have:

\(\displaystyle 2^{\dfrac{1}{n+1}}<1+\frac{1}{n}\)

or:

\(\displaystyle n2^{\dfrac{1}{n+1}}<n+1\)

This means, going back to our induction, we may now state:

\(\displaystyle 2^{\sum\limits_{j=2}^{k+1}\left(\dfrac{1}{j}\right)}<k\cdot2^{\dfrac{1}{k+1}}<k+1\)

or:

\(\displaystyle 2^{\sum\limits_{j=2}^{k+1}\left(\dfrac{1}{j}\right)}<k+1\)

We have derived $P_{k+1}$ from $P_k$, thereby completing the proof by induction.
 

FAQ: Proving $2^{\frac{1}{2}+\frac{1}{3}+\dots+\frac{1}{n}}<n$ for All $n\ge 2$

What does the inequality $2^{\frac{1}{2}+\frac{1}{3}+\dots+\frac{1}{n}}

The inequality $2^{\frac{1}{2}+\frac{1}{3}+\dots+\frac{1}{n}}

Why is it important to prove this inequality?

This inequality is important because it has several applications in mathematics and physics, such as in the study of series convergence and the analysis of algorithms.

How can we prove this inequality for all values of $n\ge 2$?

We can prove this inequality using mathematical induction, by showing that it holds for the base case $n=2$ and then assuming it holds for some arbitrary value of $n$ and proving it for the next value $n+1$.

What is the significance of the exponent $\frac{1}{2}+\frac{1}{3}+\dots+\frac{1}{n}$ in the inequality?

The exponent $\frac{1}{2}+\frac{1}{3}+\dots+\frac{1}{n}$ represents the sum of the reciprocals of the first $n$ positive integers. This sum is commonly known as the harmonic series and it is known to diverge as $n$ approaches infinity, making the inequality $2^{\frac{1}{2}+\frac{1}{3}+\dots+\frac{1}{n}}

Are there any known counterexamples to this inequality?

No, there are no known counterexamples to this inequality. In fact, it has been proven to be true for all values of $n\ge 2$ using mathematical induction.

Similar threads

Back
Top