##\lim_{n \to \infty} ## for the sequence at ##a_n##

Click For Summary

Homework Help Overview

The discussion revolves around the limit of a sequence defined recursively, where the first term is given as 3 and subsequent terms are generated using the formula ##a_{n+1} = \frac{2}{3} a_n + \frac{1}{4}##. Participants are exploring the nature of the sequence and its convergence.

Discussion Character

  • Exploratory, Conceptual clarification, Assumption checking

Approaches and Questions Raised

  • Participants are attempting to derive a formula for the sequence and are questioning whether it is necessary to classify the sequence as geometric or arithmetic. There are discussions about the implications of assuming the sequence approaches a limit and how to demonstrate that a limit exists.

Discussion Status

The discussion is ongoing, with various participants providing insights and suggestions on how to approach the problem. Some have proposed expanding the recursive definition to identify patterns, while others are clarifying the notation and concepts involved in limits.

Contextual Notes

There is some confusion regarding the notation and the distinction between limits as n approaches infinity versus other variables. Participants are also grappling with the implications of their findings and the nature of the sequence's growth.

Helly123
Messages
581
Reaction score
20

Homework Statement


##a_1## = 3
##a_{n+1}## = ##\frac{2}{3} a_n + \frac{1}{4} ##

Homework Equations


[/B]

The Attempt at a Solution


Sequence i got :
## a_1, a_2, a_3, a_4, a_5 ##
## 3, \frac{9}{4}, \frac{7}{4}, \frac{17}{12}, \frac{43}{36} ##

I tried to find the formula of ##a_n##
##a_n = \frac{3}{2} a_{n+1} - \frac{3}{8} ##

Can i get a clue? Is it necessary to know if it is geometric or arithmetic sequence?
 
Physics news on Phys.org
Helly123 said:

Homework Statement


##a_1## = 3
##a_{n+1}## = ##\frac{2}{3} a_n + \frac{1}{4} ##

Homework Equations


[/B]

The Attempt at a Solution


Sequence i got :
## a_1, a_2, a_3, a_4, a_5 ##
## 3, \frac{9}{4}, \frac{7}{4}, \frac{17}{12}, \frac{43}{36} ##

I tried to find the formula of ##a_n##
##a_n = \frac{3}{2} a_{n+1} - \frac{3}{8} ##

Can i get a clue? Is it necessary to know if it is geometric or arithmetic sequence?

It's neither arithmetic nor geometric. But you don't need to know that to find the limit. IF ##a_n## approaches a limit ##L##, then what do you get if you take the limit of ##a_{n+1}## = ##\frac{2}{3} a_n + \frac{1}{4} ##? What must ##L## be? But that's ASSUMING that there is a limit. Proving there IS a limit takes a different kind of thinking. How might you show that?
 
  • Like
Likes   Reactions: PeroK and Helly123
Helly123 said:

Homework Statement


##a_1## = 3
##a_{n+1}## = ##\frac{2}{3} a_n + \frac{1}{4} ##

Homework Equations


[/B]

The Attempt at a Solution


Sequence i got :
## a_1, a_2, a_3, a_4, a_5 ##
## 3, \frac{9}{4}, \frac{7}{4}, \frac{17}{12}, \frac{43}{36} ##

I tried to find the formula of ##a_n##
##a_n = \frac{3}{2} a_{n+1} - \frac{3}{8} ##

Can i get a clue? Is it necessary to know if it is geometric or arithmetic sequence?
Do you know what makes the difference between the two?
What did you get for ##a_{n+1}=a_{n+1}(a_{n-1})=a_{n+1}(a_{n-2})= \ldots ?##
 
  • Like
Likes   Reactions: Helly123
Helly123 said:
I tried to find the formula of ##a_n##
##a_n = \frac{3}{2} a_{n+1} - \frac{3}{8} ##
This isn't what is meant by finding the formula for ##a_n##. In your formula, to get ##a_n## you need to know ##a_{n+1}##. "Finding the formula" means being able to determine ##a_n## in terms of ##a_1##.

fresh_42 said:
Do you know what makes the difference between the two? What did you get for ##a_{n+1}=a_{n+1}(a_{n-1})=a_{n+1}(a_{n-2})= \ldots ?##
I don't know what to make of this suggestion.
 
  • Like
Likes   Reactions: Helly123
Here's a start:

##a_1 = 3##
##a_2 = \frac 2 3 a_1 + \frac 1 4##
##a_3 = \frac 2 3 (\frac 2 3 a_1 + \frac 1 4) + \frac 1 4##
The part in parentheses is ##a_2##, in terms of ##a_1##
##a_4 = \frac 2 3 ( \frac 2 3 (\frac 2 3 a_1 + \frac 1 4) + \frac 1 4) + \frac 1 4##
The part in the innermost parentheses is ##a_2##, in terms of ##a_1##.
I found it helpful to expand (multiply out) the expressions on the right.
Continue with this process, until you can see a pattern for ##a_n##.
 
  • Like
Likes   Reactions: Helly123
Mark44 said:
I don't know what to make of this suggestion.
The notation ##f=f(x)## reads: ##f## as a function of ##x##. So ##a_{n+1}=a_{n+1}(a_{n-1})## means ##a_{n+1}## as a function of ##a_{n-1}##. The same what you wrote, just downwards than upwards.
 
  • Like
Likes   Reactions: Helly123
fresh_42 said:
The notation ##f=f(x)## reads: ##f## as a function of ##x##. So ##a_{n+1}=a_{n+1}(a_{n-1})## means ##a_{n+1}## as a function of ##a_{n-1}##. The same what you wrote, just downwards than upwards.
I get function notation, but I didn't realize you were using function notation on a term of a sequence. I thought you were multiplying ##a_{n+1}## and ##a_{n-1}##.
 
Mark44 said:
I get function notation, but I didn't realize you were using function notation on a term of a sequence. I thought you were multiplying ##a_{n+1}## and ##a_{n-1}##.
Sorry for laziness. And I hoped a bit the OP would have asked to get the process going.
 
Dick said:
IF ##a_n## approaches a limit ##L##, then what do you get if you take the limit of ##a_{n+1}## = ##\frac{2}{3} a_n + \frac{1}{4} ##?
##a_n## approaches a limit ##L##. What does it mean? Lim of ##a_n## the same as limit ##L##? Or ##a_n## has value of ##L##?
And take the limit as x approaches ##\infty## ?
 
  • #10
Helly123 said:
##a_n## approaches a limit ##L##. What does it mean? Lim of ##a_n## the same as limit ##L##?
Yes, although the full expression for this is ##\lim_{n \to \infty} a_n = L##.
Helly123 said:
Or ##a_n## has value of ##L##?
No, not necessarily. ##a_n## can have L as a limit without any element of the sequence actually equalling L.
Helly123 said:
And take the limit as x approaches ##\infty## ?
No, as n approaches ##\infty##.
 
  • #11
fresh_42 said:
Do you know what makes the difference between the two?
What did you get for ##a_{n+1}=a_{n+1}(a_{n-1})=a_{n+1}(a_{n-2})= \ldots ?##
Mark44 said:
Here's a start:

##a_1 = 3##
##a_2 = \frac 2 3 a_1 + \frac 1 4##
##a_3 = \frac 2 3 (\frac 2 3 a_1 + \frac 1 4) + \frac 1 4##
The part in parentheses is ##a_2##, in terms of ##a_1##
##a_4 = \frac 2 3 ( \frac 2 3 (\frac 2 3 a_1 + \frac 1 4) + \frac 1 4) + \frac 1 4##
The part in the innermost parentheses is ##a_2##, in terms of ##a_1##.
I found it helpful to expand (multiply out) the expressions on the right.
Continue with this process, until you can see a pattern for ##a_n##.
I guess, the formula for ##a_n## will be ##(\frac{2}{3})^{(n-1)}a_1 + (n-1)\frac{1}{4}## = ##3.(\frac{2}{3})^{(n-1)} + \frac{1}{4}(n-1)##

Take the limit for ##a_n## as x approaches ##\infty## i don't get what the question wants...

Usually taking the limit as x approaches ##\infty## , i have to substitute the x variable into ##\infty##
 
  • #12
Helly123 said:
And take the limit as x approaches ##\infty## ?
Think of it as ##\frac{1}{n} \longrightarrow 0##. This sequence approaches zero, i.e. we can get arbitrary close to zero without ever reaching it.
 
  • #13
Sorry. IT HAS TO BE ##\lim_{n \to \infty} ## ALL THIS TIME. NOT ##\lim_{x \to \infty} ## . i made mistake in the question
 
  • #14
Helly123 said:
I guess, the formula for ##a_n## will be ##(\frac{2}{3})^{(n-1)}a_1 + (n-1)\frac{1}{4}## = ##3.(\frac{2}{3})^{(n-1)} + \frac{1}{4}(n-1)##
To guess is a bad idea. How often will you guess, because this first guess is wrong?
Take the limit for ##a_n## as x approaches ##\infty## i don't get what the question wants...
You have written something with ##(\frac{2}{3})^n##. So what's ##\lim_{n \to \infty} (\frac{2}{3})^n\,##?
Usually taking the limit as x approaches ##\infty## , i have to substitute the x variable into ##\infty##
There is no variable ##x##. We only have a counter ##n##. Think of the two examples I gave you.
And write down a few more terms of the sequence to see, why ##n\cdot \frac{1}{4}## is wrong.
 
  • Like
Likes   Reactions: Helly123
  • #15
Mark44 said:
Yes, although the full expression for this is ##\lim_{n \to \infty} a_n = L##.
why on that expression, ##\lim_{n \to \infty} a_n = L## and NOT ##\lim_{n \to \infty} a_n = \lim_{n \to \infty} L##
 
  • #16
Helly123 said:
I guess, the formula for ##a_n## will be ##(\frac{2}{3})^{(n-1)}a_1 + (n-1)\frac{1}{4}## = ##3.(\frac{2}{3})^{(n-1)} + \frac{1}{4}(n-1)##
No, not even close. If you take the limit of the above, as n grows large, the result approaches infinity. You worked out several terms in the sequence. Does it seem that they are growing largerl without bound. Take a look at my suggestion in post #5 again.
Helly123 said:
Take the limit for ##a_n## as x approaches ##\infty## i don't get what the question wants...
What you wrote is meaningless, since ##a_n## doesn't have anything to do with x. You want to find out what ##a_n## is doing as n gets large, not x.
Helly123 said:
Usually taking the limit as x approaches ##\infty## , i have to substitute the x variable into ##\infty##
There is no x in this problem.
 
  • Like
Likes   Reactions: Helly123
  • #17
Helly123 said:
why on that expression, ##\lim_{n \to \infty} a_n = L## and NOT ##\lim_{n \to \infty} a_n = \lim_{n \to \infty} L##
##\lim_{n \to \infty} L## is just L. Since L doesn't depend on n, it's redundant to write "lim... L". L is just a number.
 
  • #18
Helly123 said:
why on that expression, ##\lim_{n \to \infty} a_n = L## and NOT ##\lim_{n \to \infty} a_n = \lim_{n \to \infty} L##
Because ##\lim_{n \to \infty} L = L##. The limit of the sequence ##(L,L,L,L,L,L,L,L,L,L,L,L,L,L,...)## is surprisingly ##L##.
 
  • Like
Likes   Reactions: Helly123
  • #19
fresh_42 said:
Do you know what makes the difference between the two?
What did you get for ##a_{n+1}=a_{n+1}(a_{n-1})=a_{n+1}(a_{n-2})= \ldots ?##
I don't understand.. difference between ##a_{n+1}(a_{n-1})## and ##a_{n+1}(a_{n-2})## ?
The difference is changing according to n.
##a_{n+1}(a_{n-1})##
means, i fill a at (n-1) into the formula ##a_{n+1}## right?
Then why the value of ##a_{n+1}(a_{n-1})## = ##a_{n+1}(a_{n-2})## ? Since ##a_{n-1}## not the same as ##a_{n-2}##

fresh_42 said:
You have written something with ##(\frac{2}{3})^n##. So what's ##\lim_{n \to \infty} (\frac{2}{3})^n\,##?

And write down a few more terms of the sequence to see, why ##n\cdot \frac{1}{4}## is wrong.
I tried this
##a_2 - a_1## =
##-\frac{1}{3}a_{1} + \frac{1}{4}##
Then
##a_3 - a_2## = ##-\frac{1}{3}a_{2} + \frac{1}{4}##
I still have to think again
 
  • #20
Helly123 said:
I don't understand.. difference between ##a_{n+1}(a_{n-1})## and ##a_{n+1}(a_{n-2})## ?
The difference is changing according to n.
##a_{n+1}(a_{n-1})##
means, i fill a at (n-1) into the formula ##a_{n+1}## right?
Then why the value of ##a_{n+1}(a_{n-1})## = ##a_{n+1}(a_{n-2})## ? Since ##a_{n-1}## not the same as ##a_{n-2}##I tried this
##a_2 - a_1## =
##-\frac{1}{3}a_{1} + \frac{1}{4}##
Then
##a_3 - a_2## = ##-\frac{1}{3}a_{2} + \frac{1}{4}##
I still have to think again
You can as well start from the top. But in any case you will need the law of distribution, because we want to arrive at a function with only ##n## and numbers in it: ##a_n = f(n)##
 
  • #21
Mark44 said:
No, not even close. If you take the limit of the above, as n grows large, the result approaches infinity. You worked out several terms in the sequence. Does it seem that they are growing largerl without bound. Take a look at my suggestion in post #5 again.
The sequence getting smaller as n approaches ##\infty## but the formula i guessed is getting the bigger value as n approaches ##\infty## is that what you meant? That is why it's wrong?
 
  • #22
fresh_42 said:
You can as well start from the top. But in any case you will need the law of distribution, because we want to arrive at a function with only ##n## and numbers in it: ##a_n = f(n)##
Yes. But what do you mean by that 2 examples.. as what i explained in #19. I don't get that
 
  • #23
Helly123 said:
The sequence getting smaller as n approaches ##\infty## but the formula i guessed is getting the bigger value as n approaches ##\infty## is that what you meant? That is why it's wrong?
Yes, that's the obvious problem with your guess, but I also have calculated the limit and compared the results directly.
I just started with
$$
a_n = \frac{2}{3}a_{n-1}+\frac{1}{4}= \frac{2}{3}\left(\frac{2}{3}a_{n-2}+\frac{1}{4}\right)+\frac{1}{4}=\frac{2}{3}\left(\frac{2}{3}\left(\frac{2}{3}a_{n-3}+\frac{1}{4}\right)+\frac{1}{4}\right)+\frac{1}{4} = \ldots
$$
Of course I calculated this to get rid of the parentheses.
 
  • Like
Likes   Reactions: Helly123
  • #24
Helly123 said:
Yes. But what do you mean by that 2 examples.. as what i explained in #19. I don't get that
That was only to explain what a limit with ##n \to \infty## means, and that we do not have to have the limit itself in the sequence.
 
  • #25
fresh_42 said:
Yes, that's the obvious problem with your guess, but I also have calculated the limit and compared the results directly.
I just started with
$$
a_n = \frac{2}{3}a_{n-1}+\frac{1}{4}= \frac{2}{3}\left(\frac{2}{3}a_{n-2}+\frac{1}{4}\right)+\frac{1}{4}=\frac{2}{3}\left(\frac{2}{3}\left(\frac{2}{3}a_{n-3}+\frac{1}{4}\right)+\frac{1}{4}\right)+\frac{1}{4} = \ldots
$$
Of course I calculated this to get rid of the parentheses.
Is there any clue to get the ##a_n## formula with only n and numbers in it?

$$
a_\infty = \frac{2}{3}a_{(\infty-1)}+\frac{1}{4} $$
 
  • #26
Helly123 said:
Is there any clue to get the ##a_n## formula with only n and numbers in it?

$$
a_\infty = \frac{2}{3}a_{(\infty-1)}+\frac{1}{4} $$
You can't do this. For one thing, you can't have an index of ##\infty##. You also can't have ##\infty## as part of an arithmetic expression like you have in ##\infty - 1##.
fresh_42 and I are approaching this from two different directions: in post #23, he's working from ##a_{n+1}## down (i.e., starting with ##a_{n+1}## in terms of ##a_n##, then with ##a_{n+1}## in terms of ##a_{n-1}##, then with ##a_{n+1}## in terms of ##a_{n-2}##, and so on. In post #5, I working my way from ##a_2## in terms of ##a_1##, then to ##a_3## in terms of ##a_1##, and then ##a_4## in terms of ##a_1##. Do this as many times as necessary until you see a pattern.
 
  • #27
Helly123 said:
Is there any clue to get the ##a_n## formula with only n and numbers in it?
Yes. Calculate the expressions:
##a_n = \frac{2}{3}a_{n-1}+\frac{1}{4}## and ##a_{n-1} = \frac{2}{3}a_{n-2}+\frac{1}{4}## result in
$$a_n=\frac{2}{3}\left(\frac{2}{3}a_{n-2}+\frac{1}{4}\right)+\frac{1}{4}=\left(\frac{2}{3}\right)^2 a_{n-2}+\frac{1}{4}\left(\left(\frac{2}{3}\right)^0+ \left(\frac{2}{3}\right)^1 \right)$$
and in the next step with ##a_{n-2}= \frac{2}{3}a_{n-3}+\frac{1}{4}##
$$a_n=\frac{2}{3}\left(\frac{2}{3}\left(\frac{2}{3}a_{n-3}+\frac{1}{4}\right)+\frac{1}{4}\right)+\frac{1}{4}=\left(\frac{2}{3}\right)^3 a_{n-3}+\frac{1}{4}\left(\left(\frac{2}{3}\right)^0+ \left(\frac{2}{3}\right)^1 + \left(\frac{2}{3}\right)^2 \right)$$
Now look at the pattern, go on until ##a_n=\ldots \, \cdot a_1 + \ldots ##, substitute ##a_1=3## and compute ##\sum_{i=0}^n q^n## with ##q=\frac{2}{3}##. This sum can be calculated by ##(x-1)^n=(x-1)\cdot (x^{n-1}+x^{n-2}+\ldots + x^2+x+1)##.
 
  • Like
Likes   Reactions: Helly123
  • #28
fresh_42 said:
Yes. Calculate the expressions:
##a_n = \frac{2}{3}a_{n-1}+\frac{1}{4}## and ##a_{n-1} = \frac{2}{3}a_{n-2}+\frac{1}{4}## result in
$$a_n=\frac{2}{3}\left(\frac{2}{3}a_{n-2}+\frac{1}{4}\right)+\frac{1}{4}=\left(\frac{2}{3}\right)^2 a_{n-2}+\frac{1}{4}\left(\left(\frac{2}{3}\right)^0+ \left(\frac{2}{3}\right)^1 \right)$$
and in the next step with ##a_{n-2}= \frac{2}{3}a_{n-3}+\frac{1}{4}##
$$a_n=\frac{2}{3}\left(\frac{2}{3}\left(\frac{2}{3}a_{n-3}+\frac{1}{4}\right)+\frac{1}{4}\right)+\frac{1}{4}=\left(\frac{2}{3}\right)^3 a_{n-3}+\frac{1}{4}\left(\left(\frac{2}{3}\right)^0+ \left(\frac{2}{3}\right)^1 + \left(\frac{2}{3}\right)^2 \right)$$
Now look at the pattern, go on until ##a_n=\ldots \, \cdot a_1 + \ldots ##, substitute ##a_1=3## and compute ##\sum_{i=0}^n q^n## with ##q=\frac{2}{3}##. This sum can be calculated by ##(x-1)^n=(x-1)\cdot (x^{n-1}+x^{n-2}+\ldots + x^2+x+1)##.
I will try to think on it
 
  • #29
Mark44 said:
You can't do this. For one thing, you can't have an index of ##\infty##. You also can't have ##\infty## as part of an arithmetic expression like you have in ##\infty - 1##.
fresh_42 and I are approaching this from two different directions: in post #23, he's working from ##a_{n+1}## down (i.e., starting with ##a_{n+1}## in terms of ##a_n##, then with ##a_{n+1}## in terms of ##a_{n-1}##, then with ##a_{n+1}## in terms of ##a_{n-2}##, and so on. In post #5, I working my way from ##a_2## in terms of ##a_1##, then to ##a_3## in terms of ##a_1##, and then ##a_4## in terms of ##a_1##. Do this as many times as necessary until you see a pattern.
Ok. I can't use ##\infty## on sequence formula because it just not appropriate mathematically?
 
  • #30
Helly123 said:
Ok. I can't use ##\infty## on sequence formula because it just not appropriate mathematically?
Nobody uses this. It could only mean ##a_\infty = \lim_{n \to \infty}a_n## which is less ambiguous and more precise. There is no element ##a_\infty## in the sequence: each element of the sequence belongs to an index ##n##. They move on to infinity, but each one of them is at the unique ##n-##th place in the sequence. And if you will have your formula for ##a_n##, then you cannot simply short cut the limit process by writing ##\infty##, because there are no calculation rules for ##\infty##. It might happen that it sometimes leads to the same result, but I wouldn't bet on it.
 
  • Like
Likes   Reactions: Helly123

Similar threads

  • · Replies 8 ·
Replies
8
Views
2K
  • · Replies 1 ·
Replies
1
Views
2K
Replies
8
Views
3K
  • · Replies 3 ·
Replies
3
Views
1K
  • · Replies 5 ·
Replies
5
Views
1K
  • · Replies 2 ·
Replies
2
Views
2K
  • · Replies 1 ·
Replies
1
Views
1K
  • · Replies 7 ·
Replies
7
Views
2K
Replies
12
Views
2K
  • · Replies 12 ·
Replies
12
Views
2K