Showing that, for $|a|>3$ and $ngeq1$, the function $f(z)=e^z-az^n$ has exactly $n$ roots (all simple and...












2












$begingroup$



This question already has an answer here:




  • Show that if $a>e$, the equation $az^n=e^z$ admit $n$ roots in the unit disk - Rouché theorem

    1 answer





Let $|a|>3$, $ngeq1$, $ninmathbb{N}$,
then the function $$f(z)=e^z-az^n$$
has exactly $n$ roots in the disc ${zmid|z|<1}$, and that they are all simple.



Hint: look at $f(z)-f'(z)$




I am not sure how do I even approach proving existence of roots, rather then proving maybe that the function is not vanishing (and then it has logarithm, and root, but how do I show existence of n roots?)



I thought also of maybe using somehow Rouche's Theorem, since:



$$f(z)-f'(z)=az^{n-1}cdot(n-z)$$



and from there, that showing that $f$ has $n$ roots in the disk.



Thank you!










share|cite|improve this question











$endgroup$



marked as duplicate by Martin R, Adrian Keister, Abcd, Shaun, Martin Sleziak Jan 21 at 15:26


This question has been asked before and already has an answer. If those answers do not fully address your question, please ask a new question.


















  • $begingroup$
    Are you familiar with Rouche's Theorem?
    $endgroup$
    – Yiorgos S. Smyrlis
    Jan 21 at 11:24






  • 4




    $begingroup$
    Possible duplicate of Show that if $a>e$, the equation $az^n=e^z$ admit $n$ roots in the unit disk - Rouché theorem
    $endgroup$
    – Martin R
    Jan 21 at 12:46


















2












$begingroup$



This question already has an answer here:




  • Show that if $a>e$, the equation $az^n=e^z$ admit $n$ roots in the unit disk - Rouché theorem

    1 answer





Let $|a|>3$, $ngeq1$, $ninmathbb{N}$,
then the function $$f(z)=e^z-az^n$$
has exactly $n$ roots in the disc ${zmid|z|<1}$, and that they are all simple.



Hint: look at $f(z)-f'(z)$




I am not sure how do I even approach proving existence of roots, rather then proving maybe that the function is not vanishing (and then it has logarithm, and root, but how do I show existence of n roots?)



I thought also of maybe using somehow Rouche's Theorem, since:



$$f(z)-f'(z)=az^{n-1}cdot(n-z)$$



and from there, that showing that $f$ has $n$ roots in the disk.



Thank you!










share|cite|improve this question











$endgroup$



marked as duplicate by Martin R, Adrian Keister, Abcd, Shaun, Martin Sleziak Jan 21 at 15:26


This question has been asked before and already has an answer. If those answers do not fully address your question, please ask a new question.


















  • $begingroup$
    Are you familiar with Rouche's Theorem?
    $endgroup$
    – Yiorgos S. Smyrlis
    Jan 21 at 11:24






  • 4




    $begingroup$
    Possible duplicate of Show that if $a>e$, the equation $az^n=e^z$ admit $n$ roots in the unit disk - Rouché theorem
    $endgroup$
    – Martin R
    Jan 21 at 12:46
















2












2








2





$begingroup$



This question already has an answer here:




  • Show that if $a>e$, the equation $az^n=e^z$ admit $n$ roots in the unit disk - Rouché theorem

    1 answer





Let $|a|>3$, $ngeq1$, $ninmathbb{N}$,
then the function $$f(z)=e^z-az^n$$
has exactly $n$ roots in the disc ${zmid|z|<1}$, and that they are all simple.



Hint: look at $f(z)-f'(z)$




I am not sure how do I even approach proving existence of roots, rather then proving maybe that the function is not vanishing (and then it has logarithm, and root, but how do I show existence of n roots?)



I thought also of maybe using somehow Rouche's Theorem, since:



$$f(z)-f'(z)=az^{n-1}cdot(n-z)$$



and from there, that showing that $f$ has $n$ roots in the disk.



Thank you!










share|cite|improve this question











$endgroup$





This question already has an answer here:




  • Show that if $a>e$, the equation $az^n=e^z$ admit $n$ roots in the unit disk - Rouché theorem

    1 answer





Let $|a|>3$, $ngeq1$, $ninmathbb{N}$,
then the function $$f(z)=e^z-az^n$$
has exactly $n$ roots in the disc ${zmid|z|<1}$, and that they are all simple.



Hint: look at $f(z)-f'(z)$




I am not sure how do I even approach proving existence of roots, rather then proving maybe that the function is not vanishing (and then it has logarithm, and root, but how do I show existence of n roots?)



I thought also of maybe using somehow Rouche's Theorem, since:



$$f(z)-f'(z)=az^{n-1}cdot(n-z)$$



and from there, that showing that $f$ has $n$ roots in the disk.



Thank you!





This question already has an answer here:




  • Show that if $a>e$, the equation $az^n=e^z$ admit $n$ roots in the unit disk - Rouché theorem

    1 answer








complex-analysis complex-numbers roots holomorphic-functions rouches-theorem






share|cite|improve this question















share|cite|improve this question













share|cite|improve this question




share|cite|improve this question








edited Jan 21 at 12:48







user3708158

















asked Jan 21 at 11:15









user3708158user3708158

1327




1327




marked as duplicate by Martin R, Adrian Keister, Abcd, Shaun, Martin Sleziak Jan 21 at 15:26


This question has been asked before and already has an answer. If those answers do not fully address your question, please ask a new question.









marked as duplicate by Martin R, Adrian Keister, Abcd, Shaun, Martin Sleziak Jan 21 at 15:26


This question has been asked before and already has an answer. If those answers do not fully address your question, please ask a new question.














  • $begingroup$
    Are you familiar with Rouche's Theorem?
    $endgroup$
    – Yiorgos S. Smyrlis
    Jan 21 at 11:24






  • 4




    $begingroup$
    Possible duplicate of Show that if $a>e$, the equation $az^n=e^z$ admit $n$ roots in the unit disk - Rouché theorem
    $endgroup$
    – Martin R
    Jan 21 at 12:46




















  • $begingroup$
    Are you familiar with Rouche's Theorem?
    $endgroup$
    – Yiorgos S. Smyrlis
    Jan 21 at 11:24






  • 4




    $begingroup$
    Possible duplicate of Show that if $a>e$, the equation $az^n=e^z$ admit $n$ roots in the unit disk - Rouché theorem
    $endgroup$
    – Martin R
    Jan 21 at 12:46


















$begingroup$
Are you familiar with Rouche's Theorem?
$endgroup$
– Yiorgos S. Smyrlis
Jan 21 at 11:24




$begingroup$
Are you familiar with Rouche's Theorem?
$endgroup$
– Yiorgos S. Smyrlis
Jan 21 at 11:24




4




4




$begingroup$
Possible duplicate of Show that if $a>e$, the equation $az^n=e^z$ admit $n$ roots in the unit disk - Rouché theorem
$endgroup$
– Martin R
Jan 21 at 12:46






$begingroup$
Possible duplicate of Show that if $a>e$, the equation $az^n=e^z$ admit $n$ roots in the unit disk - Rouché theorem
$endgroup$
– Martin R
Jan 21 at 12:46












2 Answers
2






active

oldest

votes


















1












$begingroup$

Proving existence of roots by using Rouché's Theorem is a good idea. Let $g(z)=az^n$ then for $|z|=1$,
$$|g(z)|=|a|>3>|e^z|=e^{cos(t)}=|f(z)-g(z)|$$
where $tin [0,2pi)$.
What may we conclude?



Moreover if $|w|<1$ and $f(w)=e^w-aw^n=0$ then $w$ is NOT simple iff
$$0=f'(w)=e^w-anw^{n-1}=aw^{n-1}(w-n)$$
which holds iff $w=0$ which is not a root because $f(w)=1$. So $w$ is simple.






share|cite|improve this answer











$endgroup$













  • $begingroup$
    From the first part, I can say that $g(z)=az^n$ and $g(z)+e^z=f(z)$ has the same number of zeros in the unit Disk. from here I undersand what is going on.. I think I have problem understand the question. proving f has n roots means that there are n different $zin D$ s.t $(f(z))^{1/n}=z$ or $f(z)=0$ ? And thank you for your answer!
    $endgroup$
    – user3708158
    Jan 21 at 11:31












  • $begingroup$
    ...and $az^n$ has $n$ roots inside the unit disc, i.e. $0$ with multiplicity $n$.
    $endgroup$
    – Robert Z
    Jan 21 at 11:33










  • $begingroup$
    First you prove that they are $n$ and then, by proving that they are simple, you show that they are distinct.
    $endgroup$
    – Robert Z
    Jan 21 at 11:34










  • $begingroup$
    why is w simple iff ....? (the condition in the answer)
    $endgroup$
    – user3708158
    Jan 21 at 11:49










  • $begingroup$
    @user3708158 Sorry, my fault. See my edit.
    $endgroup$
    – Robert Z
    Jan 21 at 12:39



















1












$begingroup$

The fast way is with Rouche's Theorem.



Alternatively, set $f_t(z)=az^n-te^z$.



Observe that, $f_t(z)ne 0$, for $|z|=1$, and $tin [0,1]$, and that the
$$
g(t)=frac{1}{2pi i}int_{|z|=1}frac{f_t'(z),dz}{f_t(z)}=text{number of roots of $f$ is ${|z|<1}$} in mathbb Z
$$

Meanwhile, $g$ is continuous on $[0,1]$, and hence constant. Thus
$$
text{number of roots of $f$ is ${|z|<1}$}=frac{1}{2pi i}int_{|z|=1}frac{f'(z),dz}{f(z)}=g(1) \=g(0)=frac{1}{2pi i}int_{|z|=1}frac{nz^{n-1},dz}{z^n}=n.
$$






share|cite|improve this answer









$endgroup$













  • $begingroup$
    Why is $g(t)$ is the number of roots of f? From the argument principle, g(t) is the number of zeros of f minus the number of poles in f (including their order). why is it equal to the number of roots?
    $endgroup$
    – user3708158
    Jan 21 at 11:38










  • $begingroup$
    Clearly, $f_t$ DOES NOT have any poles or other singularities in the disk, as it is entire analytic.
    $endgroup$
    – Yiorgos S. Smyrlis
    Jan 21 at 11:41










  • $begingroup$
    Yes, in this case it is equal to the number of zeros with their order. why is it the number of roots? (I know, I am confused with the definitions)
    $endgroup$
    – user3708158
    Jan 21 at 11:42






  • 1




    $begingroup$
    Zeros and roots is the same thing, in Complex Analysis.
    $endgroup$
    – Yiorgos S. Smyrlis
    Jan 21 at 11:43






  • 1




    $begingroup$
    @user3708158 : If there is made a distinction, zeros are the zero-value points of a function, while roots are the solutions of an equation. In the arabic original the words for "(square, cubic,...) roots" and "solution" had the same abbreviation, and so Fibonacci in "Liber abaci" translated them the same, which is since traditional.
    $endgroup$
    – LutzL
    Jan 22 at 8:58


















2 Answers
2






active

oldest

votes








2 Answers
2






active

oldest

votes









active

oldest

votes






active

oldest

votes









1












$begingroup$

Proving existence of roots by using Rouché's Theorem is a good idea. Let $g(z)=az^n$ then for $|z|=1$,
$$|g(z)|=|a|>3>|e^z|=e^{cos(t)}=|f(z)-g(z)|$$
where $tin [0,2pi)$.
What may we conclude?



Moreover if $|w|<1$ and $f(w)=e^w-aw^n=0$ then $w$ is NOT simple iff
$$0=f'(w)=e^w-anw^{n-1}=aw^{n-1}(w-n)$$
which holds iff $w=0$ which is not a root because $f(w)=1$. So $w$ is simple.






share|cite|improve this answer











$endgroup$













  • $begingroup$
    From the first part, I can say that $g(z)=az^n$ and $g(z)+e^z=f(z)$ has the same number of zeros in the unit Disk. from here I undersand what is going on.. I think I have problem understand the question. proving f has n roots means that there are n different $zin D$ s.t $(f(z))^{1/n}=z$ or $f(z)=0$ ? And thank you for your answer!
    $endgroup$
    – user3708158
    Jan 21 at 11:31












  • $begingroup$
    ...and $az^n$ has $n$ roots inside the unit disc, i.e. $0$ with multiplicity $n$.
    $endgroup$
    – Robert Z
    Jan 21 at 11:33










  • $begingroup$
    First you prove that they are $n$ and then, by proving that they are simple, you show that they are distinct.
    $endgroup$
    – Robert Z
    Jan 21 at 11:34










  • $begingroup$
    why is w simple iff ....? (the condition in the answer)
    $endgroup$
    – user3708158
    Jan 21 at 11:49










  • $begingroup$
    @user3708158 Sorry, my fault. See my edit.
    $endgroup$
    – Robert Z
    Jan 21 at 12:39
















1












$begingroup$

Proving existence of roots by using Rouché's Theorem is a good idea. Let $g(z)=az^n$ then for $|z|=1$,
$$|g(z)|=|a|>3>|e^z|=e^{cos(t)}=|f(z)-g(z)|$$
where $tin [0,2pi)$.
What may we conclude?



Moreover if $|w|<1$ and $f(w)=e^w-aw^n=0$ then $w$ is NOT simple iff
$$0=f'(w)=e^w-anw^{n-1}=aw^{n-1}(w-n)$$
which holds iff $w=0$ which is not a root because $f(w)=1$. So $w$ is simple.






share|cite|improve this answer











$endgroup$













  • $begingroup$
    From the first part, I can say that $g(z)=az^n$ and $g(z)+e^z=f(z)$ has the same number of zeros in the unit Disk. from here I undersand what is going on.. I think I have problem understand the question. proving f has n roots means that there are n different $zin D$ s.t $(f(z))^{1/n}=z$ or $f(z)=0$ ? And thank you for your answer!
    $endgroup$
    – user3708158
    Jan 21 at 11:31












  • $begingroup$
    ...and $az^n$ has $n$ roots inside the unit disc, i.e. $0$ with multiplicity $n$.
    $endgroup$
    – Robert Z
    Jan 21 at 11:33










  • $begingroup$
    First you prove that they are $n$ and then, by proving that they are simple, you show that they are distinct.
    $endgroup$
    – Robert Z
    Jan 21 at 11:34










  • $begingroup$
    why is w simple iff ....? (the condition in the answer)
    $endgroup$
    – user3708158
    Jan 21 at 11:49










  • $begingroup$
    @user3708158 Sorry, my fault. See my edit.
    $endgroup$
    – Robert Z
    Jan 21 at 12:39














1












1








1





$begingroup$

Proving existence of roots by using Rouché's Theorem is a good idea. Let $g(z)=az^n$ then for $|z|=1$,
$$|g(z)|=|a|>3>|e^z|=e^{cos(t)}=|f(z)-g(z)|$$
where $tin [0,2pi)$.
What may we conclude?



Moreover if $|w|<1$ and $f(w)=e^w-aw^n=0$ then $w$ is NOT simple iff
$$0=f'(w)=e^w-anw^{n-1}=aw^{n-1}(w-n)$$
which holds iff $w=0$ which is not a root because $f(w)=1$. So $w$ is simple.






share|cite|improve this answer











$endgroup$



Proving existence of roots by using Rouché's Theorem is a good idea. Let $g(z)=az^n$ then for $|z|=1$,
$$|g(z)|=|a|>3>|e^z|=e^{cos(t)}=|f(z)-g(z)|$$
where $tin [0,2pi)$.
What may we conclude?



Moreover if $|w|<1$ and $f(w)=e^w-aw^n=0$ then $w$ is NOT simple iff
$$0=f'(w)=e^w-anw^{n-1}=aw^{n-1}(w-n)$$
which holds iff $w=0$ which is not a root because $f(w)=1$. So $w$ is simple.







share|cite|improve this answer














share|cite|improve this answer



share|cite|improve this answer








edited Jan 21 at 12:39

























answered Jan 21 at 11:20









Robert ZRobert Z

98.6k1068139




98.6k1068139












  • $begingroup$
    From the first part, I can say that $g(z)=az^n$ and $g(z)+e^z=f(z)$ has the same number of zeros in the unit Disk. from here I undersand what is going on.. I think I have problem understand the question. proving f has n roots means that there are n different $zin D$ s.t $(f(z))^{1/n}=z$ or $f(z)=0$ ? And thank you for your answer!
    $endgroup$
    – user3708158
    Jan 21 at 11:31












  • $begingroup$
    ...and $az^n$ has $n$ roots inside the unit disc, i.e. $0$ with multiplicity $n$.
    $endgroup$
    – Robert Z
    Jan 21 at 11:33










  • $begingroup$
    First you prove that they are $n$ and then, by proving that they are simple, you show that they are distinct.
    $endgroup$
    – Robert Z
    Jan 21 at 11:34










  • $begingroup$
    why is w simple iff ....? (the condition in the answer)
    $endgroup$
    – user3708158
    Jan 21 at 11:49










  • $begingroup$
    @user3708158 Sorry, my fault. See my edit.
    $endgroup$
    – Robert Z
    Jan 21 at 12:39


















  • $begingroup$
    From the first part, I can say that $g(z)=az^n$ and $g(z)+e^z=f(z)$ has the same number of zeros in the unit Disk. from here I undersand what is going on.. I think I have problem understand the question. proving f has n roots means that there are n different $zin D$ s.t $(f(z))^{1/n}=z$ or $f(z)=0$ ? And thank you for your answer!
    $endgroup$
    – user3708158
    Jan 21 at 11:31












  • $begingroup$
    ...and $az^n$ has $n$ roots inside the unit disc, i.e. $0$ with multiplicity $n$.
    $endgroup$
    – Robert Z
    Jan 21 at 11:33










  • $begingroup$
    First you prove that they are $n$ and then, by proving that they are simple, you show that they are distinct.
    $endgroup$
    – Robert Z
    Jan 21 at 11:34










  • $begingroup$
    why is w simple iff ....? (the condition in the answer)
    $endgroup$
    – user3708158
    Jan 21 at 11:49










  • $begingroup$
    @user3708158 Sorry, my fault. See my edit.
    $endgroup$
    – Robert Z
    Jan 21 at 12:39
















$begingroup$
From the first part, I can say that $g(z)=az^n$ and $g(z)+e^z=f(z)$ has the same number of zeros in the unit Disk. from here I undersand what is going on.. I think I have problem understand the question. proving f has n roots means that there are n different $zin D$ s.t $(f(z))^{1/n}=z$ or $f(z)=0$ ? And thank you for your answer!
$endgroup$
– user3708158
Jan 21 at 11:31






$begingroup$
From the first part, I can say that $g(z)=az^n$ and $g(z)+e^z=f(z)$ has the same number of zeros in the unit Disk. from here I undersand what is going on.. I think I have problem understand the question. proving f has n roots means that there are n different $zin D$ s.t $(f(z))^{1/n}=z$ or $f(z)=0$ ? And thank you for your answer!
$endgroup$
– user3708158
Jan 21 at 11:31














$begingroup$
...and $az^n$ has $n$ roots inside the unit disc, i.e. $0$ with multiplicity $n$.
$endgroup$
– Robert Z
Jan 21 at 11:33




$begingroup$
...and $az^n$ has $n$ roots inside the unit disc, i.e. $0$ with multiplicity $n$.
$endgroup$
– Robert Z
Jan 21 at 11:33












$begingroup$
First you prove that they are $n$ and then, by proving that they are simple, you show that they are distinct.
$endgroup$
– Robert Z
Jan 21 at 11:34




$begingroup$
First you prove that they are $n$ and then, by proving that they are simple, you show that they are distinct.
$endgroup$
– Robert Z
Jan 21 at 11:34












$begingroup$
why is w simple iff ....? (the condition in the answer)
$endgroup$
– user3708158
Jan 21 at 11:49




$begingroup$
why is w simple iff ....? (the condition in the answer)
$endgroup$
– user3708158
Jan 21 at 11:49












$begingroup$
@user3708158 Sorry, my fault. See my edit.
$endgroup$
– Robert Z
Jan 21 at 12:39




$begingroup$
@user3708158 Sorry, my fault. See my edit.
$endgroup$
– Robert Z
Jan 21 at 12:39











1












$begingroup$

The fast way is with Rouche's Theorem.



Alternatively, set $f_t(z)=az^n-te^z$.



Observe that, $f_t(z)ne 0$, for $|z|=1$, and $tin [0,1]$, and that the
$$
g(t)=frac{1}{2pi i}int_{|z|=1}frac{f_t'(z),dz}{f_t(z)}=text{number of roots of $f$ is ${|z|<1}$} in mathbb Z
$$

Meanwhile, $g$ is continuous on $[0,1]$, and hence constant. Thus
$$
text{number of roots of $f$ is ${|z|<1}$}=frac{1}{2pi i}int_{|z|=1}frac{f'(z),dz}{f(z)}=g(1) \=g(0)=frac{1}{2pi i}int_{|z|=1}frac{nz^{n-1},dz}{z^n}=n.
$$






share|cite|improve this answer









$endgroup$













  • $begingroup$
    Why is $g(t)$ is the number of roots of f? From the argument principle, g(t) is the number of zeros of f minus the number of poles in f (including their order). why is it equal to the number of roots?
    $endgroup$
    – user3708158
    Jan 21 at 11:38










  • $begingroup$
    Clearly, $f_t$ DOES NOT have any poles or other singularities in the disk, as it is entire analytic.
    $endgroup$
    – Yiorgos S. Smyrlis
    Jan 21 at 11:41










  • $begingroup$
    Yes, in this case it is equal to the number of zeros with their order. why is it the number of roots? (I know, I am confused with the definitions)
    $endgroup$
    – user3708158
    Jan 21 at 11:42






  • 1




    $begingroup$
    Zeros and roots is the same thing, in Complex Analysis.
    $endgroup$
    – Yiorgos S. Smyrlis
    Jan 21 at 11:43






  • 1




    $begingroup$
    @user3708158 : If there is made a distinction, zeros are the zero-value points of a function, while roots are the solutions of an equation. In the arabic original the words for "(square, cubic,...) roots" and "solution" had the same abbreviation, and so Fibonacci in "Liber abaci" translated them the same, which is since traditional.
    $endgroup$
    – LutzL
    Jan 22 at 8:58
















1












$begingroup$

The fast way is with Rouche's Theorem.



Alternatively, set $f_t(z)=az^n-te^z$.



Observe that, $f_t(z)ne 0$, for $|z|=1$, and $tin [0,1]$, and that the
$$
g(t)=frac{1}{2pi i}int_{|z|=1}frac{f_t'(z),dz}{f_t(z)}=text{number of roots of $f$ is ${|z|<1}$} in mathbb Z
$$

Meanwhile, $g$ is continuous on $[0,1]$, and hence constant. Thus
$$
text{number of roots of $f$ is ${|z|<1}$}=frac{1}{2pi i}int_{|z|=1}frac{f'(z),dz}{f(z)}=g(1) \=g(0)=frac{1}{2pi i}int_{|z|=1}frac{nz^{n-1},dz}{z^n}=n.
$$






share|cite|improve this answer









$endgroup$













  • $begingroup$
    Why is $g(t)$ is the number of roots of f? From the argument principle, g(t) is the number of zeros of f minus the number of poles in f (including their order). why is it equal to the number of roots?
    $endgroup$
    – user3708158
    Jan 21 at 11:38










  • $begingroup$
    Clearly, $f_t$ DOES NOT have any poles or other singularities in the disk, as it is entire analytic.
    $endgroup$
    – Yiorgos S. Smyrlis
    Jan 21 at 11:41










  • $begingroup$
    Yes, in this case it is equal to the number of zeros with their order. why is it the number of roots? (I know, I am confused with the definitions)
    $endgroup$
    – user3708158
    Jan 21 at 11:42






  • 1




    $begingroup$
    Zeros and roots is the same thing, in Complex Analysis.
    $endgroup$
    – Yiorgos S. Smyrlis
    Jan 21 at 11:43






  • 1




    $begingroup$
    @user3708158 : If there is made a distinction, zeros are the zero-value points of a function, while roots are the solutions of an equation. In the arabic original the words for "(square, cubic,...) roots" and "solution" had the same abbreviation, and so Fibonacci in "Liber abaci" translated them the same, which is since traditional.
    $endgroup$
    – LutzL
    Jan 22 at 8:58














1












1








1





$begingroup$

The fast way is with Rouche's Theorem.



Alternatively, set $f_t(z)=az^n-te^z$.



Observe that, $f_t(z)ne 0$, for $|z|=1$, and $tin [0,1]$, and that the
$$
g(t)=frac{1}{2pi i}int_{|z|=1}frac{f_t'(z),dz}{f_t(z)}=text{number of roots of $f$ is ${|z|<1}$} in mathbb Z
$$

Meanwhile, $g$ is continuous on $[0,1]$, and hence constant. Thus
$$
text{number of roots of $f$ is ${|z|<1}$}=frac{1}{2pi i}int_{|z|=1}frac{f'(z),dz}{f(z)}=g(1) \=g(0)=frac{1}{2pi i}int_{|z|=1}frac{nz^{n-1},dz}{z^n}=n.
$$






share|cite|improve this answer









$endgroup$



The fast way is with Rouche's Theorem.



Alternatively, set $f_t(z)=az^n-te^z$.



Observe that, $f_t(z)ne 0$, for $|z|=1$, and $tin [0,1]$, and that the
$$
g(t)=frac{1}{2pi i}int_{|z|=1}frac{f_t'(z),dz}{f_t(z)}=text{number of roots of $f$ is ${|z|<1}$} in mathbb Z
$$

Meanwhile, $g$ is continuous on $[0,1]$, and hence constant. Thus
$$
text{number of roots of $f$ is ${|z|<1}$}=frac{1}{2pi i}int_{|z|=1}frac{f'(z),dz}{f(z)}=g(1) \=g(0)=frac{1}{2pi i}int_{|z|=1}frac{nz^{n-1},dz}{z^n}=n.
$$







share|cite|improve this answer












share|cite|improve this answer



share|cite|improve this answer










answered Jan 21 at 11:31









Yiorgos S. SmyrlisYiorgos S. Smyrlis

63.4k1385163




63.4k1385163












  • $begingroup$
    Why is $g(t)$ is the number of roots of f? From the argument principle, g(t) is the number of zeros of f minus the number of poles in f (including their order). why is it equal to the number of roots?
    $endgroup$
    – user3708158
    Jan 21 at 11:38










  • $begingroup$
    Clearly, $f_t$ DOES NOT have any poles or other singularities in the disk, as it is entire analytic.
    $endgroup$
    – Yiorgos S. Smyrlis
    Jan 21 at 11:41










  • $begingroup$
    Yes, in this case it is equal to the number of zeros with their order. why is it the number of roots? (I know, I am confused with the definitions)
    $endgroup$
    – user3708158
    Jan 21 at 11:42






  • 1




    $begingroup$
    Zeros and roots is the same thing, in Complex Analysis.
    $endgroup$
    – Yiorgos S. Smyrlis
    Jan 21 at 11:43






  • 1




    $begingroup$
    @user3708158 : If there is made a distinction, zeros are the zero-value points of a function, while roots are the solutions of an equation. In the arabic original the words for "(square, cubic,...) roots" and "solution" had the same abbreviation, and so Fibonacci in "Liber abaci" translated them the same, which is since traditional.
    $endgroup$
    – LutzL
    Jan 22 at 8:58


















  • $begingroup$
    Why is $g(t)$ is the number of roots of f? From the argument principle, g(t) is the number of zeros of f minus the number of poles in f (including their order). why is it equal to the number of roots?
    $endgroup$
    – user3708158
    Jan 21 at 11:38










  • $begingroup$
    Clearly, $f_t$ DOES NOT have any poles or other singularities in the disk, as it is entire analytic.
    $endgroup$
    – Yiorgos S. Smyrlis
    Jan 21 at 11:41










  • $begingroup$
    Yes, in this case it is equal to the number of zeros with their order. why is it the number of roots? (I know, I am confused with the definitions)
    $endgroup$
    – user3708158
    Jan 21 at 11:42






  • 1




    $begingroup$
    Zeros and roots is the same thing, in Complex Analysis.
    $endgroup$
    – Yiorgos S. Smyrlis
    Jan 21 at 11:43






  • 1




    $begingroup$
    @user3708158 : If there is made a distinction, zeros are the zero-value points of a function, while roots are the solutions of an equation. In the arabic original the words for "(square, cubic,...) roots" and "solution" had the same abbreviation, and so Fibonacci in "Liber abaci" translated them the same, which is since traditional.
    $endgroup$
    – LutzL
    Jan 22 at 8:58
















$begingroup$
Why is $g(t)$ is the number of roots of f? From the argument principle, g(t) is the number of zeros of f minus the number of poles in f (including their order). why is it equal to the number of roots?
$endgroup$
– user3708158
Jan 21 at 11:38




$begingroup$
Why is $g(t)$ is the number of roots of f? From the argument principle, g(t) is the number of zeros of f minus the number of poles in f (including their order). why is it equal to the number of roots?
$endgroup$
– user3708158
Jan 21 at 11:38












$begingroup$
Clearly, $f_t$ DOES NOT have any poles or other singularities in the disk, as it is entire analytic.
$endgroup$
– Yiorgos S. Smyrlis
Jan 21 at 11:41




$begingroup$
Clearly, $f_t$ DOES NOT have any poles or other singularities in the disk, as it is entire analytic.
$endgroup$
– Yiorgos S. Smyrlis
Jan 21 at 11:41












$begingroup$
Yes, in this case it is equal to the number of zeros with their order. why is it the number of roots? (I know, I am confused with the definitions)
$endgroup$
– user3708158
Jan 21 at 11:42




$begingroup$
Yes, in this case it is equal to the number of zeros with their order. why is it the number of roots? (I know, I am confused with the definitions)
$endgroup$
– user3708158
Jan 21 at 11:42




1




1




$begingroup$
Zeros and roots is the same thing, in Complex Analysis.
$endgroup$
– Yiorgos S. Smyrlis
Jan 21 at 11:43




$begingroup$
Zeros and roots is the same thing, in Complex Analysis.
$endgroup$
– Yiorgos S. Smyrlis
Jan 21 at 11:43




1




1




$begingroup$
@user3708158 : If there is made a distinction, zeros are the zero-value points of a function, while roots are the solutions of an equation. In the arabic original the words for "(square, cubic,...) roots" and "solution" had the same abbreviation, and so Fibonacci in "Liber abaci" translated them the same, which is since traditional.
$endgroup$
– LutzL
Jan 22 at 8:58




$begingroup$
@user3708158 : If there is made a distinction, zeros are the zero-value points of a function, while roots are the solutions of an equation. In the arabic original the words for "(square, cubic,...) roots" and "solution" had the same abbreviation, and so Fibonacci in "Liber abaci" translated them the same, which is since traditional.
$endgroup$
– LutzL
Jan 22 at 8:58



Popular posts from this blog

Mario Kart Wii

What does “Dominus providebit” mean?

Antonio Litta Visconti Arese