Obtaining expression for recursive sequence












0












$begingroup$


Can someone suggest how to obtain an expression for $S[i]$ given that S[0] = 0,



$S[i]=S[i-1]*(1-gamma_i)^2 + gamma_i^2$ where $gamma_i = frac{g+1}{g+i}$



EDIT: $g>0$ and $g$ can be assumed to be a natural number.



An exact expression or a lower bound would be helpful.



Thanks!










share|cite|improve this question











$endgroup$












  • $begingroup$
    I can show an upper bound of $s_i < i/gamma_i^2$
    $endgroup$
    – John B
    Dec 5 '18 at 20:18










  • $begingroup$
    Assuming $g$>0.
    $endgroup$
    – John B
    Dec 5 '18 at 20:19
















0












$begingroup$


Can someone suggest how to obtain an expression for $S[i]$ given that S[0] = 0,



$S[i]=S[i-1]*(1-gamma_i)^2 + gamma_i^2$ where $gamma_i = frac{g+1}{g+i}$



EDIT: $g>0$ and $g$ can be assumed to be a natural number.



An exact expression or a lower bound would be helpful.



Thanks!










share|cite|improve this question











$endgroup$












  • $begingroup$
    I can show an upper bound of $s_i < i/gamma_i^2$
    $endgroup$
    – John B
    Dec 5 '18 at 20:18










  • $begingroup$
    Assuming $g$>0.
    $endgroup$
    – John B
    Dec 5 '18 at 20:19














0












0








0





$begingroup$


Can someone suggest how to obtain an expression for $S[i]$ given that S[0] = 0,



$S[i]=S[i-1]*(1-gamma_i)^2 + gamma_i^2$ where $gamma_i = frac{g+1}{g+i}$



EDIT: $g>0$ and $g$ can be assumed to be a natural number.



An exact expression or a lower bound would be helpful.



Thanks!










share|cite|improve this question











$endgroup$




Can someone suggest how to obtain an expression for $S[i]$ given that S[0] = 0,



$S[i]=S[i-1]*(1-gamma_i)^2 + gamma_i^2$ where $gamma_i = frac{g+1}{g+i}$



EDIT: $g>0$ and $g$ can be assumed to be a natural number.



An exact expression or a lower bound would be helpful.



Thanks!







ordinary-differential-equations discrete-mathematics numerical-methods numerical-calculus






share|cite|improve this question















share|cite|improve this question













share|cite|improve this question




share|cite|improve this question








edited Dec 5 '18 at 21:14







zero

















asked Dec 5 '18 at 19:29









zerozero

708




708












  • $begingroup$
    I can show an upper bound of $s_i < i/gamma_i^2$
    $endgroup$
    – John B
    Dec 5 '18 at 20:18










  • $begingroup$
    Assuming $g$>0.
    $endgroup$
    – John B
    Dec 5 '18 at 20:19


















  • $begingroup$
    I can show an upper bound of $s_i < i/gamma_i^2$
    $endgroup$
    – John B
    Dec 5 '18 at 20:18










  • $begingroup$
    Assuming $g$>0.
    $endgroup$
    – John B
    Dec 5 '18 at 20:19
















$begingroup$
I can show an upper bound of $s_i < i/gamma_i^2$
$endgroup$
– John B
Dec 5 '18 at 20:18




$begingroup$
I can show an upper bound of $s_i < i/gamma_i^2$
$endgroup$
– John B
Dec 5 '18 at 20:18












$begingroup$
Assuming $g$>0.
$endgroup$
– John B
Dec 5 '18 at 20:19




$begingroup$
Assuming $g$>0.
$endgroup$
– John B
Dec 5 '18 at 20:19










2 Answers
2






active

oldest

votes


















1












$begingroup$

I was able to determine a formula for this based off of looking at the first few elements of the sequence.



I have that $$s_i=gamma_i^2(sum_{n=1}^{i-1}(prod _{k=n}^{i-1}frac{k^2}{(g+k)^2})+1).$$



Now let’s look at bounding it from below. From your hunch that there is a $c$ such that $(g+i)s_i>c$, we proceed by finding which $c$ might give us an induction argument.



Suppose $(g+i)s_i>c$ for $i>0$. Then $$(g+i+1)s_{i+1}=frac{i^2 s_i}{(g+i+1)}+frac{(g+1)^2}{g+i+1} > frac{i^2 c+(g+1)^2}{(g+i+1)^2}.$$



If we find a $c$ such that the final expression of the previous paragraph is greater than $c$ for all $i>0$, then we have the desired result. After some algebra and working backwards, it can be shown that any $c<(g+1)/2$ satisfies it. Following through with the induction argument, we can see that the base case ($i=1$) is satisfied since $s_1>1/2$.






share|cite|improve this answer











$endgroup$













  • $begingroup$
    Thanks! Can we simplify this further to get a lower bound? In particular, can we bound $sum_{n=1}^{i-1} Pi_{k=n}^{i-1}frac{k^2}{(g+k)^2} geq c i$ where $c$ is a constant.
    $endgroup$
    – zero
    Dec 5 '18 at 21:17












  • $begingroup$
    What are we using this lower bound for? I believe that the limit as $i$ goes to infinity of $s_i$ is 0.
    $endgroup$
    – John B
    Dec 5 '18 at 21:42










  • $begingroup$
    If any lower bound will do, then 0 is a lower bound.
    $endgroup$
    – John B
    Dec 5 '18 at 21:43










  • $begingroup$
    Yes! I also believe that $S[i]$ goes to $0$ as $i rightarrow infty$ but I think $S[i] > c/(g+i)$ (i.e, a lower bound which depends on i). I tried coding it up and $S[i]*(g+i)$ seems to converge to a constant. I don't know how to prove it though.
    $endgroup$
    – zero
    Dec 5 '18 at 23:06



















1












$begingroup$

For a sequence $(S_i)_{iinmathbb N}$ generated as $S_i=a_iS_{i-1}+b_i$ with $a_i,b_iinmathbb R$, by recursively expanding once ore twice you can easily obtain a general formula.
$$begin{align}
S_i
{}={} &
a_iS_{i-1}+b_i
\
{}={} &
a_i(a_{i-1}S_{i-2}+b_{i-1})+b_i
\
& {}vdots{}
\
{}={} &
a_ia_{i-1}cdots a_1 S_0
{}+{}
b_i+a_ib_{i-1}+a_ia_{i-1}b_{i-2}+ldots+ a_ia_{i-2}cdots a_2b_1
\
{}={} &
S_0prod_{j=1}^ia_j
{}+{}
sum_{j=1}^ib_jprod_{k=j+1}^ia_k.
end{align}$$

In your case $S_0=0$, hence the first term above vanishes.
What you need is a way to compute the remaining products and sums.



Since
$
a_k
{}={}
left(1-frac{g+1}{g+k}right)^2
{}={}
frac{(k-1)^2}{(g+k)^2}
$

and
$b_j=frac{(g+1)^2}{(g+j)^2}$, you have
$$begin{align}
color{red}{b_j}prod_{k=j+1}^ia_k
{}={} &
color{red}{frac{(g+1)^2}{(g+j)^2}}
,,
frac{j^2}{(g+j+1)^2}
frac{(j+1)^2}{(g+j+2)^2}
{}cdots{}
frac{(i-2)^2}{(g+i-1)^2}
frac{(i-1)^2}{(g+i)^2}
\
{}={} &
color{red}{frac{(g+1)^2}{(g+j)^2}}
,,
frac{(i-1)!^2}{(j-1)!^2}
frac{(g+j)!^2}{(g+i)!^2}
\
{}={} &
(g+1)^2
frac{(i-1)!^2}{(j-1)!^2}
frac{(g+j-1)!^2}{(g+i)!^2},
end{align}$$

where for simplicity $g$ was assumed natural.
The general case can easily be adapted by redefining the factorial accordingly (that is, $g!:=g$ and $(x+1)!:=(x+1)x!$, defined for $xin g+mathbb N$).



Therefore,
$$begin{align}
S_i
{}={} &
frac{(g+1)^2(i-1)!^2}{(g+i)!^2}
sum_{j=0}^{i-1}
frac{(g+j)!^2}{j!^2}
\
tag{1}
mbox{(EDIT)}quad
{}={} &
frac{1}{binom{g+i}{i-1}^2}
sum_{j=0}^{i-1}
binom{g+j}{j}^2.
end{align}$$

Now, in order to obtain a lower bound, notice that
$$
binom{g+j}{j}
{}={}
binom{g+j-1}{j-1}frac{g+j}{j}
{}geq{}
binom{g+j-1}{j-1}left(textstyle 1+frac{g}{i-1}right)
quad
mbox{for }jleq i-1
$$

therefore,
$$
binom{g+j}{j}^2
{}geq{}
binom{g}{0}^2left(textstyle 1+frac{g}{i-1}right)^{2j}
{}={}
left(textstyle 1+frac{g}{i-1}right)^{2j},
$$

and plugging this into (1) yields
$$begin{align}
S_i
{}geq{} &
frac{1}{binom{g+i}{i-1}^2}
sum_{j=0}^{i-1}
left(textstyle 1+frac{g}{i-1}right)^{2j}
{}={}
frac{1}{binom{g+i}{i-1}^2}
frac{(1+frac{g}{i-1})^{2i}-1}{(1+frac{g}{i-1})^2-1}.
end{align}$$






share|cite|improve this answer











$endgroup$













  • $begingroup$
    Thanks for the answer! Would it be possible to get something like $S_i(g+i) geq c $ from here, where $c$ is a constant.
    $endgroup$
    – zero
    Dec 5 '18 at 21:13










  • $begingroup$
    I added a lower bound, I don't know if that's what you need though
    $endgroup$
    – AndreasT
    Dec 6 '18 at 10:42











Your Answer





StackExchange.ifUsing("editor", function () {
return StackExchange.using("mathjaxEditing", function () {
StackExchange.MarkdownEditor.creationCallbacks.add(function (editor, postfix) {
StackExchange.mathjaxEditing.prepareWmdForMathJax(editor, postfix, [["$", "$"], ["\\(","\\)"]]);
});
});
}, "mathjax-editing");

StackExchange.ready(function() {
var channelOptions = {
tags: "".split(" "),
id: "69"
};
initTagRenderer("".split(" "), "".split(" "), channelOptions);

StackExchange.using("externalEditor", function() {
// Have to fire editor after snippets, if snippets enabled
if (StackExchange.settings.snippets.snippetsEnabled) {
StackExchange.using("snippets", function() {
createEditor();
});
}
else {
createEditor();
}
});

function createEditor() {
StackExchange.prepareEditor({
heartbeatType: 'answer',
autoActivateHeartbeat: false,
convertImagesToLinks: true,
noModals: true,
showLowRepImageUploadWarning: true,
reputationToPostImages: 10,
bindNavPrevention: true,
postfix: "",
imageUploader: {
brandingHtml: "Powered by u003ca class="icon-imgur-white" href="https://imgur.com/"u003eu003c/au003e",
contentPolicyHtml: "User contributions licensed under u003ca href="https://creativecommons.org/licenses/by-sa/3.0/"u003ecc by-sa 3.0 with attribution requiredu003c/au003e u003ca href="https://stackoverflow.com/legal/content-policy"u003e(content policy)u003c/au003e",
allowUrls: true
},
noCode: true, onDemand: true,
discardSelector: ".discard-answer"
,immediatelyShowMarkdownHelp:true
});


}
});














draft saved

draft discarded


















StackExchange.ready(
function () {
StackExchange.openid.initPostLogin('.new-post-login', 'https%3a%2f%2fmath.stackexchange.com%2fquestions%2f3027538%2fobtaining-expression-for-recursive-sequence%23new-answer', 'question_page');
}
);

Post as a guest















Required, but never shown

























2 Answers
2






active

oldest

votes








2 Answers
2






active

oldest

votes









active

oldest

votes






active

oldest

votes









1












$begingroup$

I was able to determine a formula for this based off of looking at the first few elements of the sequence.



I have that $$s_i=gamma_i^2(sum_{n=1}^{i-1}(prod _{k=n}^{i-1}frac{k^2}{(g+k)^2})+1).$$



Now let’s look at bounding it from below. From your hunch that there is a $c$ such that $(g+i)s_i>c$, we proceed by finding which $c$ might give us an induction argument.



Suppose $(g+i)s_i>c$ for $i>0$. Then $$(g+i+1)s_{i+1}=frac{i^2 s_i}{(g+i+1)}+frac{(g+1)^2}{g+i+1} > frac{i^2 c+(g+1)^2}{(g+i+1)^2}.$$



If we find a $c$ such that the final expression of the previous paragraph is greater than $c$ for all $i>0$, then we have the desired result. After some algebra and working backwards, it can be shown that any $c<(g+1)/2$ satisfies it. Following through with the induction argument, we can see that the base case ($i=1$) is satisfied since $s_1>1/2$.






share|cite|improve this answer











$endgroup$













  • $begingroup$
    Thanks! Can we simplify this further to get a lower bound? In particular, can we bound $sum_{n=1}^{i-1} Pi_{k=n}^{i-1}frac{k^2}{(g+k)^2} geq c i$ where $c$ is a constant.
    $endgroup$
    – zero
    Dec 5 '18 at 21:17












  • $begingroup$
    What are we using this lower bound for? I believe that the limit as $i$ goes to infinity of $s_i$ is 0.
    $endgroup$
    – John B
    Dec 5 '18 at 21:42










  • $begingroup$
    If any lower bound will do, then 0 is a lower bound.
    $endgroup$
    – John B
    Dec 5 '18 at 21:43










  • $begingroup$
    Yes! I also believe that $S[i]$ goes to $0$ as $i rightarrow infty$ but I think $S[i] > c/(g+i)$ (i.e, a lower bound which depends on i). I tried coding it up and $S[i]*(g+i)$ seems to converge to a constant. I don't know how to prove it though.
    $endgroup$
    – zero
    Dec 5 '18 at 23:06
















1












$begingroup$

I was able to determine a formula for this based off of looking at the first few elements of the sequence.



I have that $$s_i=gamma_i^2(sum_{n=1}^{i-1}(prod _{k=n}^{i-1}frac{k^2}{(g+k)^2})+1).$$



Now let’s look at bounding it from below. From your hunch that there is a $c$ such that $(g+i)s_i>c$, we proceed by finding which $c$ might give us an induction argument.



Suppose $(g+i)s_i>c$ for $i>0$. Then $$(g+i+1)s_{i+1}=frac{i^2 s_i}{(g+i+1)}+frac{(g+1)^2}{g+i+1} > frac{i^2 c+(g+1)^2}{(g+i+1)^2}.$$



If we find a $c$ such that the final expression of the previous paragraph is greater than $c$ for all $i>0$, then we have the desired result. After some algebra and working backwards, it can be shown that any $c<(g+1)/2$ satisfies it. Following through with the induction argument, we can see that the base case ($i=1$) is satisfied since $s_1>1/2$.






share|cite|improve this answer











$endgroup$













  • $begingroup$
    Thanks! Can we simplify this further to get a lower bound? In particular, can we bound $sum_{n=1}^{i-1} Pi_{k=n}^{i-1}frac{k^2}{(g+k)^2} geq c i$ where $c$ is a constant.
    $endgroup$
    – zero
    Dec 5 '18 at 21:17












  • $begingroup$
    What are we using this lower bound for? I believe that the limit as $i$ goes to infinity of $s_i$ is 0.
    $endgroup$
    – John B
    Dec 5 '18 at 21:42










  • $begingroup$
    If any lower bound will do, then 0 is a lower bound.
    $endgroup$
    – John B
    Dec 5 '18 at 21:43










  • $begingroup$
    Yes! I also believe that $S[i]$ goes to $0$ as $i rightarrow infty$ but I think $S[i] > c/(g+i)$ (i.e, a lower bound which depends on i). I tried coding it up and $S[i]*(g+i)$ seems to converge to a constant. I don't know how to prove it though.
    $endgroup$
    – zero
    Dec 5 '18 at 23:06














1












1








1





$begingroup$

I was able to determine a formula for this based off of looking at the first few elements of the sequence.



I have that $$s_i=gamma_i^2(sum_{n=1}^{i-1}(prod _{k=n}^{i-1}frac{k^2}{(g+k)^2})+1).$$



Now let’s look at bounding it from below. From your hunch that there is a $c$ such that $(g+i)s_i>c$, we proceed by finding which $c$ might give us an induction argument.



Suppose $(g+i)s_i>c$ for $i>0$. Then $$(g+i+1)s_{i+1}=frac{i^2 s_i}{(g+i+1)}+frac{(g+1)^2}{g+i+1} > frac{i^2 c+(g+1)^2}{(g+i+1)^2}.$$



If we find a $c$ such that the final expression of the previous paragraph is greater than $c$ for all $i>0$, then we have the desired result. After some algebra and working backwards, it can be shown that any $c<(g+1)/2$ satisfies it. Following through with the induction argument, we can see that the base case ($i=1$) is satisfied since $s_1>1/2$.






share|cite|improve this answer











$endgroup$



I was able to determine a formula for this based off of looking at the first few elements of the sequence.



I have that $$s_i=gamma_i^2(sum_{n=1}^{i-1}(prod _{k=n}^{i-1}frac{k^2}{(g+k)^2})+1).$$



Now let’s look at bounding it from below. From your hunch that there is a $c$ such that $(g+i)s_i>c$, we proceed by finding which $c$ might give us an induction argument.



Suppose $(g+i)s_i>c$ for $i>0$. Then $$(g+i+1)s_{i+1}=frac{i^2 s_i}{(g+i+1)}+frac{(g+1)^2}{g+i+1} > frac{i^2 c+(g+1)^2}{(g+i+1)^2}.$$



If we find a $c$ such that the final expression of the previous paragraph is greater than $c$ for all $i>0$, then we have the desired result. After some algebra and working backwards, it can be shown that any $c<(g+1)/2$ satisfies it. Following through with the induction argument, we can see that the base case ($i=1$) is satisfied since $s_1>1/2$.







share|cite|improve this answer














share|cite|improve this answer



share|cite|improve this answer








edited Dec 6 '18 at 2:41

























answered Dec 5 '18 at 20:29









John BJohn B

1937




1937












  • $begingroup$
    Thanks! Can we simplify this further to get a lower bound? In particular, can we bound $sum_{n=1}^{i-1} Pi_{k=n}^{i-1}frac{k^2}{(g+k)^2} geq c i$ where $c$ is a constant.
    $endgroup$
    – zero
    Dec 5 '18 at 21:17












  • $begingroup$
    What are we using this lower bound for? I believe that the limit as $i$ goes to infinity of $s_i$ is 0.
    $endgroup$
    – John B
    Dec 5 '18 at 21:42










  • $begingroup$
    If any lower bound will do, then 0 is a lower bound.
    $endgroup$
    – John B
    Dec 5 '18 at 21:43










  • $begingroup$
    Yes! I also believe that $S[i]$ goes to $0$ as $i rightarrow infty$ but I think $S[i] > c/(g+i)$ (i.e, a lower bound which depends on i). I tried coding it up and $S[i]*(g+i)$ seems to converge to a constant. I don't know how to prove it though.
    $endgroup$
    – zero
    Dec 5 '18 at 23:06


















  • $begingroup$
    Thanks! Can we simplify this further to get a lower bound? In particular, can we bound $sum_{n=1}^{i-1} Pi_{k=n}^{i-1}frac{k^2}{(g+k)^2} geq c i$ where $c$ is a constant.
    $endgroup$
    – zero
    Dec 5 '18 at 21:17












  • $begingroup$
    What are we using this lower bound for? I believe that the limit as $i$ goes to infinity of $s_i$ is 0.
    $endgroup$
    – John B
    Dec 5 '18 at 21:42










  • $begingroup$
    If any lower bound will do, then 0 is a lower bound.
    $endgroup$
    – John B
    Dec 5 '18 at 21:43










  • $begingroup$
    Yes! I also believe that $S[i]$ goes to $0$ as $i rightarrow infty$ but I think $S[i] > c/(g+i)$ (i.e, a lower bound which depends on i). I tried coding it up and $S[i]*(g+i)$ seems to converge to a constant. I don't know how to prove it though.
    $endgroup$
    – zero
    Dec 5 '18 at 23:06
















$begingroup$
Thanks! Can we simplify this further to get a lower bound? In particular, can we bound $sum_{n=1}^{i-1} Pi_{k=n}^{i-1}frac{k^2}{(g+k)^2} geq c i$ where $c$ is a constant.
$endgroup$
– zero
Dec 5 '18 at 21:17






$begingroup$
Thanks! Can we simplify this further to get a lower bound? In particular, can we bound $sum_{n=1}^{i-1} Pi_{k=n}^{i-1}frac{k^2}{(g+k)^2} geq c i$ where $c$ is a constant.
$endgroup$
– zero
Dec 5 '18 at 21:17














$begingroup$
What are we using this lower bound for? I believe that the limit as $i$ goes to infinity of $s_i$ is 0.
$endgroup$
– John B
Dec 5 '18 at 21:42




$begingroup$
What are we using this lower bound for? I believe that the limit as $i$ goes to infinity of $s_i$ is 0.
$endgroup$
– John B
Dec 5 '18 at 21:42












$begingroup$
If any lower bound will do, then 0 is a lower bound.
$endgroup$
– John B
Dec 5 '18 at 21:43




$begingroup$
If any lower bound will do, then 0 is a lower bound.
$endgroup$
– John B
Dec 5 '18 at 21:43












$begingroup$
Yes! I also believe that $S[i]$ goes to $0$ as $i rightarrow infty$ but I think $S[i] > c/(g+i)$ (i.e, a lower bound which depends on i). I tried coding it up and $S[i]*(g+i)$ seems to converge to a constant. I don't know how to prove it though.
$endgroup$
– zero
Dec 5 '18 at 23:06




$begingroup$
Yes! I also believe that $S[i]$ goes to $0$ as $i rightarrow infty$ but I think $S[i] > c/(g+i)$ (i.e, a lower bound which depends on i). I tried coding it up and $S[i]*(g+i)$ seems to converge to a constant. I don't know how to prove it though.
$endgroup$
– zero
Dec 5 '18 at 23:06











1












$begingroup$

For a sequence $(S_i)_{iinmathbb N}$ generated as $S_i=a_iS_{i-1}+b_i$ with $a_i,b_iinmathbb R$, by recursively expanding once ore twice you can easily obtain a general formula.
$$begin{align}
S_i
{}={} &
a_iS_{i-1}+b_i
\
{}={} &
a_i(a_{i-1}S_{i-2}+b_{i-1})+b_i
\
& {}vdots{}
\
{}={} &
a_ia_{i-1}cdots a_1 S_0
{}+{}
b_i+a_ib_{i-1}+a_ia_{i-1}b_{i-2}+ldots+ a_ia_{i-2}cdots a_2b_1
\
{}={} &
S_0prod_{j=1}^ia_j
{}+{}
sum_{j=1}^ib_jprod_{k=j+1}^ia_k.
end{align}$$

In your case $S_0=0$, hence the first term above vanishes.
What you need is a way to compute the remaining products and sums.



Since
$
a_k
{}={}
left(1-frac{g+1}{g+k}right)^2
{}={}
frac{(k-1)^2}{(g+k)^2}
$

and
$b_j=frac{(g+1)^2}{(g+j)^2}$, you have
$$begin{align}
color{red}{b_j}prod_{k=j+1}^ia_k
{}={} &
color{red}{frac{(g+1)^2}{(g+j)^2}}
,,
frac{j^2}{(g+j+1)^2}
frac{(j+1)^2}{(g+j+2)^2}
{}cdots{}
frac{(i-2)^2}{(g+i-1)^2}
frac{(i-1)^2}{(g+i)^2}
\
{}={} &
color{red}{frac{(g+1)^2}{(g+j)^2}}
,,
frac{(i-1)!^2}{(j-1)!^2}
frac{(g+j)!^2}{(g+i)!^2}
\
{}={} &
(g+1)^2
frac{(i-1)!^2}{(j-1)!^2}
frac{(g+j-1)!^2}{(g+i)!^2},
end{align}$$

where for simplicity $g$ was assumed natural.
The general case can easily be adapted by redefining the factorial accordingly (that is, $g!:=g$ and $(x+1)!:=(x+1)x!$, defined for $xin g+mathbb N$).



Therefore,
$$begin{align}
S_i
{}={} &
frac{(g+1)^2(i-1)!^2}{(g+i)!^2}
sum_{j=0}^{i-1}
frac{(g+j)!^2}{j!^2}
\
tag{1}
mbox{(EDIT)}quad
{}={} &
frac{1}{binom{g+i}{i-1}^2}
sum_{j=0}^{i-1}
binom{g+j}{j}^2.
end{align}$$

Now, in order to obtain a lower bound, notice that
$$
binom{g+j}{j}
{}={}
binom{g+j-1}{j-1}frac{g+j}{j}
{}geq{}
binom{g+j-1}{j-1}left(textstyle 1+frac{g}{i-1}right)
quad
mbox{for }jleq i-1
$$

therefore,
$$
binom{g+j}{j}^2
{}geq{}
binom{g}{0}^2left(textstyle 1+frac{g}{i-1}right)^{2j}
{}={}
left(textstyle 1+frac{g}{i-1}right)^{2j},
$$

and plugging this into (1) yields
$$begin{align}
S_i
{}geq{} &
frac{1}{binom{g+i}{i-1}^2}
sum_{j=0}^{i-1}
left(textstyle 1+frac{g}{i-1}right)^{2j}
{}={}
frac{1}{binom{g+i}{i-1}^2}
frac{(1+frac{g}{i-1})^{2i}-1}{(1+frac{g}{i-1})^2-1}.
end{align}$$






share|cite|improve this answer











$endgroup$













  • $begingroup$
    Thanks for the answer! Would it be possible to get something like $S_i(g+i) geq c $ from here, where $c$ is a constant.
    $endgroup$
    – zero
    Dec 5 '18 at 21:13










  • $begingroup$
    I added a lower bound, I don't know if that's what you need though
    $endgroup$
    – AndreasT
    Dec 6 '18 at 10:42
















1












$begingroup$

For a sequence $(S_i)_{iinmathbb N}$ generated as $S_i=a_iS_{i-1}+b_i$ with $a_i,b_iinmathbb R$, by recursively expanding once ore twice you can easily obtain a general formula.
$$begin{align}
S_i
{}={} &
a_iS_{i-1}+b_i
\
{}={} &
a_i(a_{i-1}S_{i-2}+b_{i-1})+b_i
\
& {}vdots{}
\
{}={} &
a_ia_{i-1}cdots a_1 S_0
{}+{}
b_i+a_ib_{i-1}+a_ia_{i-1}b_{i-2}+ldots+ a_ia_{i-2}cdots a_2b_1
\
{}={} &
S_0prod_{j=1}^ia_j
{}+{}
sum_{j=1}^ib_jprod_{k=j+1}^ia_k.
end{align}$$

In your case $S_0=0$, hence the first term above vanishes.
What you need is a way to compute the remaining products and sums.



Since
$
a_k
{}={}
left(1-frac{g+1}{g+k}right)^2
{}={}
frac{(k-1)^2}{(g+k)^2}
$

and
$b_j=frac{(g+1)^2}{(g+j)^2}$, you have
$$begin{align}
color{red}{b_j}prod_{k=j+1}^ia_k
{}={} &
color{red}{frac{(g+1)^2}{(g+j)^2}}
,,
frac{j^2}{(g+j+1)^2}
frac{(j+1)^2}{(g+j+2)^2}
{}cdots{}
frac{(i-2)^2}{(g+i-1)^2}
frac{(i-1)^2}{(g+i)^2}
\
{}={} &
color{red}{frac{(g+1)^2}{(g+j)^2}}
,,
frac{(i-1)!^2}{(j-1)!^2}
frac{(g+j)!^2}{(g+i)!^2}
\
{}={} &
(g+1)^2
frac{(i-1)!^2}{(j-1)!^2}
frac{(g+j-1)!^2}{(g+i)!^2},
end{align}$$

where for simplicity $g$ was assumed natural.
The general case can easily be adapted by redefining the factorial accordingly (that is, $g!:=g$ and $(x+1)!:=(x+1)x!$, defined for $xin g+mathbb N$).



Therefore,
$$begin{align}
S_i
{}={} &
frac{(g+1)^2(i-1)!^2}{(g+i)!^2}
sum_{j=0}^{i-1}
frac{(g+j)!^2}{j!^2}
\
tag{1}
mbox{(EDIT)}quad
{}={} &
frac{1}{binom{g+i}{i-1}^2}
sum_{j=0}^{i-1}
binom{g+j}{j}^2.
end{align}$$

Now, in order to obtain a lower bound, notice that
$$
binom{g+j}{j}
{}={}
binom{g+j-1}{j-1}frac{g+j}{j}
{}geq{}
binom{g+j-1}{j-1}left(textstyle 1+frac{g}{i-1}right)
quad
mbox{for }jleq i-1
$$

therefore,
$$
binom{g+j}{j}^2
{}geq{}
binom{g}{0}^2left(textstyle 1+frac{g}{i-1}right)^{2j}
{}={}
left(textstyle 1+frac{g}{i-1}right)^{2j},
$$

and plugging this into (1) yields
$$begin{align}
S_i
{}geq{} &
frac{1}{binom{g+i}{i-1}^2}
sum_{j=0}^{i-1}
left(textstyle 1+frac{g}{i-1}right)^{2j}
{}={}
frac{1}{binom{g+i}{i-1}^2}
frac{(1+frac{g}{i-1})^{2i}-1}{(1+frac{g}{i-1})^2-1}.
end{align}$$






share|cite|improve this answer











$endgroup$













  • $begingroup$
    Thanks for the answer! Would it be possible to get something like $S_i(g+i) geq c $ from here, where $c$ is a constant.
    $endgroup$
    – zero
    Dec 5 '18 at 21:13










  • $begingroup$
    I added a lower bound, I don't know if that's what you need though
    $endgroup$
    – AndreasT
    Dec 6 '18 at 10:42














1












1








1





$begingroup$

For a sequence $(S_i)_{iinmathbb N}$ generated as $S_i=a_iS_{i-1}+b_i$ with $a_i,b_iinmathbb R$, by recursively expanding once ore twice you can easily obtain a general formula.
$$begin{align}
S_i
{}={} &
a_iS_{i-1}+b_i
\
{}={} &
a_i(a_{i-1}S_{i-2}+b_{i-1})+b_i
\
& {}vdots{}
\
{}={} &
a_ia_{i-1}cdots a_1 S_0
{}+{}
b_i+a_ib_{i-1}+a_ia_{i-1}b_{i-2}+ldots+ a_ia_{i-2}cdots a_2b_1
\
{}={} &
S_0prod_{j=1}^ia_j
{}+{}
sum_{j=1}^ib_jprod_{k=j+1}^ia_k.
end{align}$$

In your case $S_0=0$, hence the first term above vanishes.
What you need is a way to compute the remaining products and sums.



Since
$
a_k
{}={}
left(1-frac{g+1}{g+k}right)^2
{}={}
frac{(k-1)^2}{(g+k)^2}
$

and
$b_j=frac{(g+1)^2}{(g+j)^2}$, you have
$$begin{align}
color{red}{b_j}prod_{k=j+1}^ia_k
{}={} &
color{red}{frac{(g+1)^2}{(g+j)^2}}
,,
frac{j^2}{(g+j+1)^2}
frac{(j+1)^2}{(g+j+2)^2}
{}cdots{}
frac{(i-2)^2}{(g+i-1)^2}
frac{(i-1)^2}{(g+i)^2}
\
{}={} &
color{red}{frac{(g+1)^2}{(g+j)^2}}
,,
frac{(i-1)!^2}{(j-1)!^2}
frac{(g+j)!^2}{(g+i)!^2}
\
{}={} &
(g+1)^2
frac{(i-1)!^2}{(j-1)!^2}
frac{(g+j-1)!^2}{(g+i)!^2},
end{align}$$

where for simplicity $g$ was assumed natural.
The general case can easily be adapted by redefining the factorial accordingly (that is, $g!:=g$ and $(x+1)!:=(x+1)x!$, defined for $xin g+mathbb N$).



Therefore,
$$begin{align}
S_i
{}={} &
frac{(g+1)^2(i-1)!^2}{(g+i)!^2}
sum_{j=0}^{i-1}
frac{(g+j)!^2}{j!^2}
\
tag{1}
mbox{(EDIT)}quad
{}={} &
frac{1}{binom{g+i}{i-1}^2}
sum_{j=0}^{i-1}
binom{g+j}{j}^2.
end{align}$$

Now, in order to obtain a lower bound, notice that
$$
binom{g+j}{j}
{}={}
binom{g+j-1}{j-1}frac{g+j}{j}
{}geq{}
binom{g+j-1}{j-1}left(textstyle 1+frac{g}{i-1}right)
quad
mbox{for }jleq i-1
$$

therefore,
$$
binom{g+j}{j}^2
{}geq{}
binom{g}{0}^2left(textstyle 1+frac{g}{i-1}right)^{2j}
{}={}
left(textstyle 1+frac{g}{i-1}right)^{2j},
$$

and plugging this into (1) yields
$$begin{align}
S_i
{}geq{} &
frac{1}{binom{g+i}{i-1}^2}
sum_{j=0}^{i-1}
left(textstyle 1+frac{g}{i-1}right)^{2j}
{}={}
frac{1}{binom{g+i}{i-1}^2}
frac{(1+frac{g}{i-1})^{2i}-1}{(1+frac{g}{i-1})^2-1}.
end{align}$$






share|cite|improve this answer











$endgroup$



For a sequence $(S_i)_{iinmathbb N}$ generated as $S_i=a_iS_{i-1}+b_i$ with $a_i,b_iinmathbb R$, by recursively expanding once ore twice you can easily obtain a general formula.
$$begin{align}
S_i
{}={} &
a_iS_{i-1}+b_i
\
{}={} &
a_i(a_{i-1}S_{i-2}+b_{i-1})+b_i
\
& {}vdots{}
\
{}={} &
a_ia_{i-1}cdots a_1 S_0
{}+{}
b_i+a_ib_{i-1}+a_ia_{i-1}b_{i-2}+ldots+ a_ia_{i-2}cdots a_2b_1
\
{}={} &
S_0prod_{j=1}^ia_j
{}+{}
sum_{j=1}^ib_jprod_{k=j+1}^ia_k.
end{align}$$

In your case $S_0=0$, hence the first term above vanishes.
What you need is a way to compute the remaining products and sums.



Since
$
a_k
{}={}
left(1-frac{g+1}{g+k}right)^2
{}={}
frac{(k-1)^2}{(g+k)^2}
$

and
$b_j=frac{(g+1)^2}{(g+j)^2}$, you have
$$begin{align}
color{red}{b_j}prod_{k=j+1}^ia_k
{}={} &
color{red}{frac{(g+1)^2}{(g+j)^2}}
,,
frac{j^2}{(g+j+1)^2}
frac{(j+1)^2}{(g+j+2)^2}
{}cdots{}
frac{(i-2)^2}{(g+i-1)^2}
frac{(i-1)^2}{(g+i)^2}
\
{}={} &
color{red}{frac{(g+1)^2}{(g+j)^2}}
,,
frac{(i-1)!^2}{(j-1)!^2}
frac{(g+j)!^2}{(g+i)!^2}
\
{}={} &
(g+1)^2
frac{(i-1)!^2}{(j-1)!^2}
frac{(g+j-1)!^2}{(g+i)!^2},
end{align}$$

where for simplicity $g$ was assumed natural.
The general case can easily be adapted by redefining the factorial accordingly (that is, $g!:=g$ and $(x+1)!:=(x+1)x!$, defined for $xin g+mathbb N$).



Therefore,
$$begin{align}
S_i
{}={} &
frac{(g+1)^2(i-1)!^2}{(g+i)!^2}
sum_{j=0}^{i-1}
frac{(g+j)!^2}{j!^2}
\
tag{1}
mbox{(EDIT)}quad
{}={} &
frac{1}{binom{g+i}{i-1}^2}
sum_{j=0}^{i-1}
binom{g+j}{j}^2.
end{align}$$

Now, in order to obtain a lower bound, notice that
$$
binom{g+j}{j}
{}={}
binom{g+j-1}{j-1}frac{g+j}{j}
{}geq{}
binom{g+j-1}{j-1}left(textstyle 1+frac{g}{i-1}right)
quad
mbox{for }jleq i-1
$$

therefore,
$$
binom{g+j}{j}^2
{}geq{}
binom{g}{0}^2left(textstyle 1+frac{g}{i-1}right)^{2j}
{}={}
left(textstyle 1+frac{g}{i-1}right)^{2j},
$$

and plugging this into (1) yields
$$begin{align}
S_i
{}geq{} &
frac{1}{binom{g+i}{i-1}^2}
sum_{j=0}^{i-1}
left(textstyle 1+frac{g}{i-1}right)^{2j}
{}={}
frac{1}{binom{g+i}{i-1}^2}
frac{(1+frac{g}{i-1})^{2i}-1}{(1+frac{g}{i-1})^2-1}.
end{align}$$







share|cite|improve this answer














share|cite|improve this answer



share|cite|improve this answer








edited Dec 6 '18 at 10:42

























answered Dec 5 '18 at 20:33









AndreasTAndreasT

3,3061324




3,3061324












  • $begingroup$
    Thanks for the answer! Would it be possible to get something like $S_i(g+i) geq c $ from here, where $c$ is a constant.
    $endgroup$
    – zero
    Dec 5 '18 at 21:13










  • $begingroup$
    I added a lower bound, I don't know if that's what you need though
    $endgroup$
    – AndreasT
    Dec 6 '18 at 10:42


















  • $begingroup$
    Thanks for the answer! Would it be possible to get something like $S_i(g+i) geq c $ from here, where $c$ is a constant.
    $endgroup$
    – zero
    Dec 5 '18 at 21:13










  • $begingroup$
    I added a lower bound, I don't know if that's what you need though
    $endgroup$
    – AndreasT
    Dec 6 '18 at 10:42
















$begingroup$
Thanks for the answer! Would it be possible to get something like $S_i(g+i) geq c $ from here, where $c$ is a constant.
$endgroup$
– zero
Dec 5 '18 at 21:13




$begingroup$
Thanks for the answer! Would it be possible to get something like $S_i(g+i) geq c $ from here, where $c$ is a constant.
$endgroup$
– zero
Dec 5 '18 at 21:13












$begingroup$
I added a lower bound, I don't know if that's what you need though
$endgroup$
– AndreasT
Dec 6 '18 at 10:42




$begingroup$
I added a lower bound, I don't know if that's what you need though
$endgroup$
– AndreasT
Dec 6 '18 at 10:42


















draft saved

draft discarded




















































Thanks for contributing an answer to Mathematics Stack Exchange!


  • Please be sure to answer the question. Provide details and share your research!

But avoid



  • Asking for help, clarification, or responding to other answers.

  • Making statements based on opinion; back them up with references or personal experience.


Use MathJax to format equations. MathJax reference.


To learn more, see our tips on writing great answers.




draft saved


draft discarded














StackExchange.ready(
function () {
StackExchange.openid.initPostLogin('.new-post-login', 'https%3a%2f%2fmath.stackexchange.com%2fquestions%2f3027538%2fobtaining-expression-for-recursive-sequence%23new-answer', 'question_page');
}
);

Post as a guest















Required, but never shown





















































Required, but never shown














Required, but never shown












Required, but never shown







Required, but never shown

































Required, but never shown














Required, but never shown












Required, but never shown







Required, but never shown







Popular posts from this blog

Biblatex bibliography style without URLs when DOI exists (in Overleaf with Zotero bibliography)

ComboBox Display Member on multiple fields

Is it possible to collect Nectar points via Trainline?