How to see that $forall tinmathbb{R} quadfrac{2}{sqrt{2pi}}(t+frac{1}{t})e^{-t^2/2} leq frac{4}{t^2}$?...
up vote
0
down vote
favorite
How to prove the following inequality $forall tinmathbb{R}$:$$frac{2}{sqrt{2pi}}(t+frac{1}{t})e^{-t^2/2} leq frac{4}{t^2}$$
calculus derivatives inequality
closed as off-topic by TheSimpliFire, user21820, user302797, Vidyanshu Mishra, Trevor Gunn Dec 1 at 16:40
This question appears to be off-topic. The users who voted to close gave this specific reason:
- "This question is missing context or other details: Please improve the question by providing additional context, which ideally includes your thoughts on the problem and any attempts you have made to solve it. This information helps others identify where you have difficulties and helps them write answers appropriate to your experience level." – TheSimpliFire, user21820, user302797, Vidyanshu Mishra, Trevor Gunn
If this question can be reworded to fit the rules in the help center, please edit the question.
add a comment |
up vote
0
down vote
favorite
How to prove the following inequality $forall tinmathbb{R}$:$$frac{2}{sqrt{2pi}}(t+frac{1}{t})e^{-t^2/2} leq frac{4}{t^2}$$
calculus derivatives inequality
closed as off-topic by TheSimpliFire, user21820, user302797, Vidyanshu Mishra, Trevor Gunn Dec 1 at 16:40
This question appears to be off-topic. The users who voted to close gave this specific reason:
- "This question is missing context or other details: Please improve the question by providing additional context, which ideally includes your thoughts on the problem and any attempts you have made to solve it. This information helps others identify where you have difficulties and helps them write answers appropriate to your experience level." – TheSimpliFire, user21820, user302797, Vidyanshu Mishra, Trevor Gunn
If this question can be reworded to fit the rules in the help center, please edit the question.
It looks like rewriting with log might help.
– Mark
Nov 24 at 9:28
3
Please write down what you have tried to solve this problem to avoid closure. Don't you think that $frac{expleft(-frac{t^2}2right)}{sqrt{2pi}}$ looks like a density function?
– TheSimpliFire
Nov 24 at 9:40
add a comment |
up vote
0
down vote
favorite
up vote
0
down vote
favorite
How to prove the following inequality $forall tinmathbb{R}$:$$frac{2}{sqrt{2pi}}(t+frac{1}{t})e^{-t^2/2} leq frac{4}{t^2}$$
calculus derivatives inequality
How to prove the following inequality $forall tinmathbb{R}$:$$frac{2}{sqrt{2pi}}(t+frac{1}{t})e^{-t^2/2} leq frac{4}{t^2}$$
calculus derivatives inequality
calculus derivatives inequality
edited Nov 24 at 9:59
gimusi
1
1
asked Nov 24 at 9:24
ShaoyuPei
1537
1537
closed as off-topic by TheSimpliFire, user21820, user302797, Vidyanshu Mishra, Trevor Gunn Dec 1 at 16:40
This question appears to be off-topic. The users who voted to close gave this specific reason:
- "This question is missing context or other details: Please improve the question by providing additional context, which ideally includes your thoughts on the problem and any attempts you have made to solve it. This information helps others identify where you have difficulties and helps them write answers appropriate to your experience level." – TheSimpliFire, user21820, user302797, Vidyanshu Mishra, Trevor Gunn
If this question can be reworded to fit the rules in the help center, please edit the question.
closed as off-topic by TheSimpliFire, user21820, user302797, Vidyanshu Mishra, Trevor Gunn Dec 1 at 16:40
This question appears to be off-topic. The users who voted to close gave this specific reason:
- "This question is missing context or other details: Please improve the question by providing additional context, which ideally includes your thoughts on the problem and any attempts you have made to solve it. This information helps others identify where you have difficulties and helps them write answers appropriate to your experience level." – TheSimpliFire, user21820, user302797, Vidyanshu Mishra, Trevor Gunn
If this question can be reworded to fit the rules in the help center, please edit the question.
It looks like rewriting with log might help.
– Mark
Nov 24 at 9:28
3
Please write down what you have tried to solve this problem to avoid closure. Don't you think that $frac{expleft(-frac{t^2}2right)}{sqrt{2pi}}$ looks like a density function?
– TheSimpliFire
Nov 24 at 9:40
add a comment |
It looks like rewriting with log might help.
– Mark
Nov 24 at 9:28
3
Please write down what you have tried to solve this problem to avoid closure. Don't you think that $frac{expleft(-frac{t^2}2right)}{sqrt{2pi}}$ looks like a density function?
– TheSimpliFire
Nov 24 at 9:40
It looks like rewriting with log might help.
– Mark
Nov 24 at 9:28
It looks like rewriting with log might help.
– Mark
Nov 24 at 9:28
3
3
Please write down what you have tried to solve this problem to avoid closure. Don't you think that $frac{expleft(-frac{t^2}2right)}{sqrt{2pi}}$ looks like a density function?
– TheSimpliFire
Nov 24 at 9:40
Please write down what you have tried to solve this problem to avoid closure. Don't you think that $frac{expleft(-frac{t^2}2right)}{sqrt{2pi}}$ looks like a density function?
– TheSimpliFire
Nov 24 at 9:40
add a comment |
3 Answers
3
active
oldest
votes
up vote
2
down vote
accepted
HINT
We have that
$$frac{2}{sqrt{2pi}}left(t+frac{1}{t}right)e^{-t^2/2} leq frac{4}{t^2} iff e^{t^2/2} geq frac{1}{2sqrt{2pi}}(t^3+t)$$
and then consider $f(t)=e^{t^2/2} -frac{1}{2sqrt{2pi}}(t^3+t)ge 0$, indeed we have
- $f'(t)=te^{t^2/2}-frac{3t^2+1}{2sqrt{2pi}}$
- $f''(t)=t^2e^{t^2/2}+e^{t^2/2}-frac{3t}{sqrt{2pi}}$
and we can show that $f'(t)$ has exactly one root at $x_0$ which is a positive minimum poinf for $f(t)$.
As an alternative, we can also use that $e^xge 1+x+frac12 x^2$ and consider the simpler
$$g(t)=1+frac12t^2+frac18t^4 -frac{1}{2sqrt{2pi}}(t^3+t)ge 0$$
show the function is increasing when t>0?
– ShaoyuPei
Nov 24 at 9:40
@ShaoyuPei We can show that $f(t) ge 0$ by its derivatives. Indeed we can show that $f'(t)$ has exactly one root and $f(t)$ has a positive absolute minimum here.
– gimusi
Nov 24 at 9:41
@ShaoyuPei To do that we can consider $f''(t)$ and show in a similar way that it is positive and then $f'(t)$ is strictly increasing.
– gimusi
Nov 24 at 9:45
thanks learn a lot
– ShaoyuPei
Nov 24 at 10:10
@ShaoyuPei That's really the most important things! You are welcome, Bye
– gimusi
Nov 24 at 10:11
add a comment |
up vote
0
down vote
(Since there are a powers of $2$)
Taking $log_2$ on both sides
$1-dfrac12-log_2 sqrt{pi}+log_2left(t+dfrac1tright)-dfrac{t^2}{2}leq2-2log_2t $ $implies dfrac{t^2}{2}-log_2left(t^3+tright)+log_2 sqrt{pi}+dfrac32geq0$
We show $f(t)=dfrac{t^2}{2}-log_2left(t^3+tright)+log_2 sqrt{pi}+dfrac32geq0 $. $f'(t)=t-dfrac{3t^2+1}{t^3+t}=0implies t^4-2t^2-1=0implies t^2-1=pmsqrt{2}implies t=pmsqrt{1pmsqrt{2}}.$
(We do not consider $pmsqrt{1-sqrt{2}}$ as t does not belong to $mathbb{C}$ and $-sqrt{1+sqrt{2}}:$ since $log_2$ cannot take negative values).
$f''(t)|_{t=sqrt{1+sqrt{2}}}=dfrac{t^6+5t^4+t^2+1}{(t^3+t)^2}>0$ and $f(t)|_{t=sqrt{1+sqrt{2}}}>0$. So the root is a point of positive minimum.
Hence the inequality holds.
add a comment |
up vote
0
down vote
We have $2sqrt{2pi} > 2sqrt{6.25} = 5$, so it is enough to prove
$5e^{t^2/2} > t + t^3$ (for all real $t$ - including $0$, for
which the original inequality doesn't make sense). This is obvious
if $t leqslant 0$, because the LHS is $> 0$ and the RHS is
$leqslant 0$. If $0 < t leqslant sqrt{2}$, then the LHS is $> 5$
and the RHS is $leqslant 3sqrt{2} < 5$, so the inequality is true. If
$sqrt{2} < t leqslant 2$, then $e^{t^2/2} > e > 2$, therefore
$5e^{t^2/2} > 10 geqslant t + t^3$, so the inequality is true
again. If $t > 2$, then from the series for $e^x$, $e^{t^2/2} > 1 + t^2/2 + t^4/8$, therefore $5e^{t^2/2} > t^2 + t^4/2 > t + t^3$. So the inequality holds
in all cases (and equality never holds).
add a comment |
3 Answers
3
active
oldest
votes
3 Answers
3
active
oldest
votes
active
oldest
votes
active
oldest
votes
up vote
2
down vote
accepted
HINT
We have that
$$frac{2}{sqrt{2pi}}left(t+frac{1}{t}right)e^{-t^2/2} leq frac{4}{t^2} iff e^{t^2/2} geq frac{1}{2sqrt{2pi}}(t^3+t)$$
and then consider $f(t)=e^{t^2/2} -frac{1}{2sqrt{2pi}}(t^3+t)ge 0$, indeed we have
- $f'(t)=te^{t^2/2}-frac{3t^2+1}{2sqrt{2pi}}$
- $f''(t)=t^2e^{t^2/2}+e^{t^2/2}-frac{3t}{sqrt{2pi}}$
and we can show that $f'(t)$ has exactly one root at $x_0$ which is a positive minimum poinf for $f(t)$.
As an alternative, we can also use that $e^xge 1+x+frac12 x^2$ and consider the simpler
$$g(t)=1+frac12t^2+frac18t^4 -frac{1}{2sqrt{2pi}}(t^3+t)ge 0$$
show the function is increasing when t>0?
– ShaoyuPei
Nov 24 at 9:40
@ShaoyuPei We can show that $f(t) ge 0$ by its derivatives. Indeed we can show that $f'(t)$ has exactly one root and $f(t)$ has a positive absolute minimum here.
– gimusi
Nov 24 at 9:41
@ShaoyuPei To do that we can consider $f''(t)$ and show in a similar way that it is positive and then $f'(t)$ is strictly increasing.
– gimusi
Nov 24 at 9:45
thanks learn a lot
– ShaoyuPei
Nov 24 at 10:10
@ShaoyuPei That's really the most important things! You are welcome, Bye
– gimusi
Nov 24 at 10:11
add a comment |
up vote
2
down vote
accepted
HINT
We have that
$$frac{2}{sqrt{2pi}}left(t+frac{1}{t}right)e^{-t^2/2} leq frac{4}{t^2} iff e^{t^2/2} geq frac{1}{2sqrt{2pi}}(t^3+t)$$
and then consider $f(t)=e^{t^2/2} -frac{1}{2sqrt{2pi}}(t^3+t)ge 0$, indeed we have
- $f'(t)=te^{t^2/2}-frac{3t^2+1}{2sqrt{2pi}}$
- $f''(t)=t^2e^{t^2/2}+e^{t^2/2}-frac{3t}{sqrt{2pi}}$
and we can show that $f'(t)$ has exactly one root at $x_0$ which is a positive minimum poinf for $f(t)$.
As an alternative, we can also use that $e^xge 1+x+frac12 x^2$ and consider the simpler
$$g(t)=1+frac12t^2+frac18t^4 -frac{1}{2sqrt{2pi}}(t^3+t)ge 0$$
show the function is increasing when t>0?
– ShaoyuPei
Nov 24 at 9:40
@ShaoyuPei We can show that $f(t) ge 0$ by its derivatives. Indeed we can show that $f'(t)$ has exactly one root and $f(t)$ has a positive absolute minimum here.
– gimusi
Nov 24 at 9:41
@ShaoyuPei To do that we can consider $f''(t)$ and show in a similar way that it is positive and then $f'(t)$ is strictly increasing.
– gimusi
Nov 24 at 9:45
thanks learn a lot
– ShaoyuPei
Nov 24 at 10:10
@ShaoyuPei That's really the most important things! You are welcome, Bye
– gimusi
Nov 24 at 10:11
add a comment |
up vote
2
down vote
accepted
up vote
2
down vote
accepted
HINT
We have that
$$frac{2}{sqrt{2pi}}left(t+frac{1}{t}right)e^{-t^2/2} leq frac{4}{t^2} iff e^{t^2/2} geq frac{1}{2sqrt{2pi}}(t^3+t)$$
and then consider $f(t)=e^{t^2/2} -frac{1}{2sqrt{2pi}}(t^3+t)ge 0$, indeed we have
- $f'(t)=te^{t^2/2}-frac{3t^2+1}{2sqrt{2pi}}$
- $f''(t)=t^2e^{t^2/2}+e^{t^2/2}-frac{3t}{sqrt{2pi}}$
and we can show that $f'(t)$ has exactly one root at $x_0$ which is a positive minimum poinf for $f(t)$.
As an alternative, we can also use that $e^xge 1+x+frac12 x^2$ and consider the simpler
$$g(t)=1+frac12t^2+frac18t^4 -frac{1}{2sqrt{2pi}}(t^3+t)ge 0$$
HINT
We have that
$$frac{2}{sqrt{2pi}}left(t+frac{1}{t}right)e^{-t^2/2} leq frac{4}{t^2} iff e^{t^2/2} geq frac{1}{2sqrt{2pi}}(t^3+t)$$
and then consider $f(t)=e^{t^2/2} -frac{1}{2sqrt{2pi}}(t^3+t)ge 0$, indeed we have
- $f'(t)=te^{t^2/2}-frac{3t^2+1}{2sqrt{2pi}}$
- $f''(t)=t^2e^{t^2/2}+e^{t^2/2}-frac{3t}{sqrt{2pi}}$
and we can show that $f'(t)$ has exactly one root at $x_0$ which is a positive minimum poinf for $f(t)$.
As an alternative, we can also use that $e^xge 1+x+frac12 x^2$ and consider the simpler
$$g(t)=1+frac12t^2+frac18t^4 -frac{1}{2sqrt{2pi}}(t^3+t)ge 0$$
edited Nov 24 at 9:55
answered Nov 24 at 9:33
gimusi
1
1
show the function is increasing when t>0?
– ShaoyuPei
Nov 24 at 9:40
@ShaoyuPei We can show that $f(t) ge 0$ by its derivatives. Indeed we can show that $f'(t)$ has exactly one root and $f(t)$ has a positive absolute minimum here.
– gimusi
Nov 24 at 9:41
@ShaoyuPei To do that we can consider $f''(t)$ and show in a similar way that it is positive and then $f'(t)$ is strictly increasing.
– gimusi
Nov 24 at 9:45
thanks learn a lot
– ShaoyuPei
Nov 24 at 10:10
@ShaoyuPei That's really the most important things! You are welcome, Bye
– gimusi
Nov 24 at 10:11
add a comment |
show the function is increasing when t>0?
– ShaoyuPei
Nov 24 at 9:40
@ShaoyuPei We can show that $f(t) ge 0$ by its derivatives. Indeed we can show that $f'(t)$ has exactly one root and $f(t)$ has a positive absolute minimum here.
– gimusi
Nov 24 at 9:41
@ShaoyuPei To do that we can consider $f''(t)$ and show in a similar way that it is positive and then $f'(t)$ is strictly increasing.
– gimusi
Nov 24 at 9:45
thanks learn a lot
– ShaoyuPei
Nov 24 at 10:10
@ShaoyuPei That's really the most important things! You are welcome, Bye
– gimusi
Nov 24 at 10:11
show the function is increasing when t>0?
– ShaoyuPei
Nov 24 at 9:40
show the function is increasing when t>0?
– ShaoyuPei
Nov 24 at 9:40
@ShaoyuPei We can show that $f(t) ge 0$ by its derivatives. Indeed we can show that $f'(t)$ has exactly one root and $f(t)$ has a positive absolute minimum here.
– gimusi
Nov 24 at 9:41
@ShaoyuPei We can show that $f(t) ge 0$ by its derivatives. Indeed we can show that $f'(t)$ has exactly one root and $f(t)$ has a positive absolute minimum here.
– gimusi
Nov 24 at 9:41
@ShaoyuPei To do that we can consider $f''(t)$ and show in a similar way that it is positive and then $f'(t)$ is strictly increasing.
– gimusi
Nov 24 at 9:45
@ShaoyuPei To do that we can consider $f''(t)$ and show in a similar way that it is positive and then $f'(t)$ is strictly increasing.
– gimusi
Nov 24 at 9:45
thanks learn a lot
– ShaoyuPei
Nov 24 at 10:10
thanks learn a lot
– ShaoyuPei
Nov 24 at 10:10
@ShaoyuPei That's really the most important things! You are welcome, Bye
– gimusi
Nov 24 at 10:11
@ShaoyuPei That's really the most important things! You are welcome, Bye
– gimusi
Nov 24 at 10:11
add a comment |
up vote
0
down vote
(Since there are a powers of $2$)
Taking $log_2$ on both sides
$1-dfrac12-log_2 sqrt{pi}+log_2left(t+dfrac1tright)-dfrac{t^2}{2}leq2-2log_2t $ $implies dfrac{t^2}{2}-log_2left(t^3+tright)+log_2 sqrt{pi}+dfrac32geq0$
We show $f(t)=dfrac{t^2}{2}-log_2left(t^3+tright)+log_2 sqrt{pi}+dfrac32geq0 $. $f'(t)=t-dfrac{3t^2+1}{t^3+t}=0implies t^4-2t^2-1=0implies t^2-1=pmsqrt{2}implies t=pmsqrt{1pmsqrt{2}}.$
(We do not consider $pmsqrt{1-sqrt{2}}$ as t does not belong to $mathbb{C}$ and $-sqrt{1+sqrt{2}}:$ since $log_2$ cannot take negative values).
$f''(t)|_{t=sqrt{1+sqrt{2}}}=dfrac{t^6+5t^4+t^2+1}{(t^3+t)^2}>0$ and $f(t)|_{t=sqrt{1+sqrt{2}}}>0$. So the root is a point of positive minimum.
Hence the inequality holds.
add a comment |
up vote
0
down vote
(Since there are a powers of $2$)
Taking $log_2$ on both sides
$1-dfrac12-log_2 sqrt{pi}+log_2left(t+dfrac1tright)-dfrac{t^2}{2}leq2-2log_2t $ $implies dfrac{t^2}{2}-log_2left(t^3+tright)+log_2 sqrt{pi}+dfrac32geq0$
We show $f(t)=dfrac{t^2}{2}-log_2left(t^3+tright)+log_2 sqrt{pi}+dfrac32geq0 $. $f'(t)=t-dfrac{3t^2+1}{t^3+t}=0implies t^4-2t^2-1=0implies t^2-1=pmsqrt{2}implies t=pmsqrt{1pmsqrt{2}}.$
(We do not consider $pmsqrt{1-sqrt{2}}$ as t does not belong to $mathbb{C}$ and $-sqrt{1+sqrt{2}}:$ since $log_2$ cannot take negative values).
$f''(t)|_{t=sqrt{1+sqrt{2}}}=dfrac{t^6+5t^4+t^2+1}{(t^3+t)^2}>0$ and $f(t)|_{t=sqrt{1+sqrt{2}}}>0$. So the root is a point of positive minimum.
Hence the inequality holds.
add a comment |
up vote
0
down vote
up vote
0
down vote
(Since there are a powers of $2$)
Taking $log_2$ on both sides
$1-dfrac12-log_2 sqrt{pi}+log_2left(t+dfrac1tright)-dfrac{t^2}{2}leq2-2log_2t $ $implies dfrac{t^2}{2}-log_2left(t^3+tright)+log_2 sqrt{pi}+dfrac32geq0$
We show $f(t)=dfrac{t^2}{2}-log_2left(t^3+tright)+log_2 sqrt{pi}+dfrac32geq0 $. $f'(t)=t-dfrac{3t^2+1}{t^3+t}=0implies t^4-2t^2-1=0implies t^2-1=pmsqrt{2}implies t=pmsqrt{1pmsqrt{2}}.$
(We do not consider $pmsqrt{1-sqrt{2}}$ as t does not belong to $mathbb{C}$ and $-sqrt{1+sqrt{2}}:$ since $log_2$ cannot take negative values).
$f''(t)|_{t=sqrt{1+sqrt{2}}}=dfrac{t^6+5t^4+t^2+1}{(t^3+t)^2}>0$ and $f(t)|_{t=sqrt{1+sqrt{2}}}>0$. So the root is a point of positive minimum.
Hence the inequality holds.
(Since there are a powers of $2$)
Taking $log_2$ on both sides
$1-dfrac12-log_2 sqrt{pi}+log_2left(t+dfrac1tright)-dfrac{t^2}{2}leq2-2log_2t $ $implies dfrac{t^2}{2}-log_2left(t^3+tright)+log_2 sqrt{pi}+dfrac32geq0$
We show $f(t)=dfrac{t^2}{2}-log_2left(t^3+tright)+log_2 sqrt{pi}+dfrac32geq0 $. $f'(t)=t-dfrac{3t^2+1}{t^3+t}=0implies t^4-2t^2-1=0implies t^2-1=pmsqrt{2}implies t=pmsqrt{1pmsqrt{2}}.$
(We do not consider $pmsqrt{1-sqrt{2}}$ as t does not belong to $mathbb{C}$ and $-sqrt{1+sqrt{2}}:$ since $log_2$ cannot take negative values).
$f''(t)|_{t=sqrt{1+sqrt{2}}}=dfrac{t^6+5t^4+t^2+1}{(t^3+t)^2}>0$ and $f(t)|_{t=sqrt{1+sqrt{2}}}>0$. So the root is a point of positive minimum.
Hence the inequality holds.
answered Nov 24 at 10:42
Yadati Kiran
1,444518
1,444518
add a comment |
add a comment |
up vote
0
down vote
We have $2sqrt{2pi} > 2sqrt{6.25} = 5$, so it is enough to prove
$5e^{t^2/2} > t + t^3$ (for all real $t$ - including $0$, for
which the original inequality doesn't make sense). This is obvious
if $t leqslant 0$, because the LHS is $> 0$ and the RHS is
$leqslant 0$. If $0 < t leqslant sqrt{2}$, then the LHS is $> 5$
and the RHS is $leqslant 3sqrt{2} < 5$, so the inequality is true. If
$sqrt{2} < t leqslant 2$, then $e^{t^2/2} > e > 2$, therefore
$5e^{t^2/2} > 10 geqslant t + t^3$, so the inequality is true
again. If $t > 2$, then from the series for $e^x$, $e^{t^2/2} > 1 + t^2/2 + t^4/8$, therefore $5e^{t^2/2} > t^2 + t^4/2 > t + t^3$. So the inequality holds
in all cases (and equality never holds).
add a comment |
up vote
0
down vote
We have $2sqrt{2pi} > 2sqrt{6.25} = 5$, so it is enough to prove
$5e^{t^2/2} > t + t^3$ (for all real $t$ - including $0$, for
which the original inequality doesn't make sense). This is obvious
if $t leqslant 0$, because the LHS is $> 0$ and the RHS is
$leqslant 0$. If $0 < t leqslant sqrt{2}$, then the LHS is $> 5$
and the RHS is $leqslant 3sqrt{2} < 5$, so the inequality is true. If
$sqrt{2} < t leqslant 2$, then $e^{t^2/2} > e > 2$, therefore
$5e^{t^2/2} > 10 geqslant t + t^3$, so the inequality is true
again. If $t > 2$, then from the series for $e^x$, $e^{t^2/2} > 1 + t^2/2 + t^4/8$, therefore $5e^{t^2/2} > t^2 + t^4/2 > t + t^3$. So the inequality holds
in all cases (and equality never holds).
add a comment |
up vote
0
down vote
up vote
0
down vote
We have $2sqrt{2pi} > 2sqrt{6.25} = 5$, so it is enough to prove
$5e^{t^2/2} > t + t^3$ (for all real $t$ - including $0$, for
which the original inequality doesn't make sense). This is obvious
if $t leqslant 0$, because the LHS is $> 0$ and the RHS is
$leqslant 0$. If $0 < t leqslant sqrt{2}$, then the LHS is $> 5$
and the RHS is $leqslant 3sqrt{2} < 5$, so the inequality is true. If
$sqrt{2} < t leqslant 2$, then $e^{t^2/2} > e > 2$, therefore
$5e^{t^2/2} > 10 geqslant t + t^3$, so the inequality is true
again. If $t > 2$, then from the series for $e^x$, $e^{t^2/2} > 1 + t^2/2 + t^4/8$, therefore $5e^{t^2/2} > t^2 + t^4/2 > t + t^3$. So the inequality holds
in all cases (and equality never holds).
We have $2sqrt{2pi} > 2sqrt{6.25} = 5$, so it is enough to prove
$5e^{t^2/2} > t + t^3$ (for all real $t$ - including $0$, for
which the original inequality doesn't make sense). This is obvious
if $t leqslant 0$, because the LHS is $> 0$ and the RHS is
$leqslant 0$. If $0 < t leqslant sqrt{2}$, then the LHS is $> 5$
and the RHS is $leqslant 3sqrt{2} < 5$, so the inequality is true. If
$sqrt{2} < t leqslant 2$, then $e^{t^2/2} > e > 2$, therefore
$5e^{t^2/2} > 10 geqslant t + t^3$, so the inequality is true
again. If $t > 2$, then from the series for $e^x$, $e^{t^2/2} > 1 + t^2/2 + t^4/8$, therefore $5e^{t^2/2} > t^2 + t^4/2 > t + t^3$. So the inequality holds
in all cases (and equality never holds).
edited Nov 24 at 13:28
answered Nov 24 at 12:53
Calum Gilhooley
4,097529
4,097529
add a comment |
add a comment |
It looks like rewriting with log might help.
– Mark
Nov 24 at 9:28
3
Please write down what you have tried to solve this problem to avoid closure. Don't you think that $frac{expleft(-frac{t^2}2right)}{sqrt{2pi}}$ looks like a density function?
– TheSimpliFire
Nov 24 at 9:40