How can I prove it without using Taylor series, L'Hopital's rule, or integrals,or series expansion? I don't know how. Please help.
First I have tried to use epsilon-delta, I also tried to use sequences but I failed.
How can I prove it without using Taylor series, L'Hopital's rule, or integrals,or series expansion? I don't know how. Please help.
First I have tried to use epsilon-delta, I also tried to use sequences but I failed.
On
Since $$ \mathop {\lim }\limits_{x \to \pm \infty } \left( {1 + \frac{1} {x}} \right)^x = e $$ you have that $$ \mathop {\lim }\limits_{x \to \pm \infty } x\log \left( {1 + \frac{1} {x}} \right) = 1 $$ Now, let be $t=\frac{1}{x}$. You have that $$ \mathop {\lim }\limits_{t \to 0} \frac{1} {t}\log \left( {1 + t} \right) = 1 $$
On
Still calculus approach, just no integrals or explicit L'Hopital:
Let $\varphi(x)=\log(x+1)-x$ for $x>0$, then $\varphi'(x)<0$ for $x>0$, so $\varphi(x)\leq\varphi(0)=0$, we get $\dfrac{\log(1+x)}{x}\leq 1$ for $x>0$.
On the other hand, let $\varphi(x)=\log(x+1)-x+x^{2}/2$ for $0<x<1$, then $\varphi'(x)>0$ for $0<x<1$, so $\dfrac{\log(x+1)}{x}\geq 1-\dfrac{1}{2}x$ for $0<x<1$.
Finally, one concludes by Squeeze Theorem.
For negative part, we exploit the trick like the following: \begin{align*} \lim_{x\rightarrow 0^{-}}\dfrac{\log(x+1)}{x}&=\lim_{y\rightarrow 0^{+}}\dfrac{\log(1-y)}{-y}\\ &=\lim_{y\rightarrow 0^{+}}\dfrac{-\log(1-y)}{y}\\ &=\lim_{y\rightarrow 0^{+}}\dfrac{\log\left(\dfrac{1}{1-y}\right)}{y}\\ &=\lim_{y\rightarrow 0^{+}}\dfrac{\log\left(1+\dfrac{y}{1-y}\right)}{y}\\ &=\lim_{y\rightarrow 0^{+}}\dfrac{\log\left(1+\dfrac{y}{1-y}\right)}{\dfrac{y}{1-y}}\dfrac{1}{1-y}\\ &=1\cdot 1\\ &=1. \end{align*}
On
The Inequalities section of Wikipedia's "List of Logarithmic Identities" page states
$$\frac{x}{\sqrt{1+x+x^2/12}} \le \ln(1 + x) \le \frac{x}{\sqrt{1+x}} \tag{1}\label{eq1A}$$
for $0 \le x$, and the reverse for $-1 \lt x \le 0$. Dividing by $x$ and then having $x \to 0^{+}$ as shown, and with the reverse inequalities for $x \to 0^{-}$, gives $1$ on either side in both cases. Thus, by the Squeeze Theorem, you have
$$\lim_{x \to 0}\frac{\log(1+x)}{x} = 1 \tag{2}\label{eq2A}$$
Hint For $x >0$ whenever when $n \leq \frac{1}{x} <n+1$ you have $$\log (1+\frac{1}{n+1})^n\leq \frac{\log(x+1)}{x} \leq \log (1+\frac{1}{n})^{n+1}$$
Use the fact that
$$\lim_n (1+\frac{1}{n+1})^n= \lim_n (1+\frac{1}{n})^{n+1}=e$$
Edit With your definition of $e$.
Hint 1: Use your definition of $e$ to show that $$\lim_{x \to 0} \frac{e^x-1}{x} =e$$
Hint 2 If you meet it, apply the formula for $(f^{-1})'(0)$. If not, calculate $$\lim_{x\rightarrow 0}\frac{\log(x+1)}{x}$$ via the substitution $x=e^y-1$. Then, you get $$\lim_{x\rightarrow 0}\frac{\log(x+1)}{x}=\lim_{y\rightarrow 0}\frac{\log(e^y-1+1)}{e^y-1}$$