Use a left and midpoint Rienman sum to approximate

  • Thread starter Thread starter flyingpig
  • Start date Start date
  • Tags Tags
    Sums
Click For Summary

Homework Help Overview

The problem involves using left and midpoint Riemann sums to approximate the integral \(\int_{0}^{3} x^3 dx\). Participants are discussing the setup of the Riemann sums and the implications of choosing different endpoints for the approximation.

Discussion Character

  • Exploratory, Conceptual clarification, Problem interpretation, Assumption checking

Approaches and Questions Raised

  • Participants are attempting to clarify the choice of the left endpoint 'a' for the Riemann sum and how to calculate the values of \(x_i\). There are questions about the relationship between the left and right endpoints and whether they will yield the same results. Some participants are exploring the implications of choosing different values for \(n\) in the context of approximating the integral.

Discussion Status

The discussion is ongoing, with various interpretations being explored regarding the setup of the Riemann sums. Some participants have offered guidance on how to approach the problem, while others are questioning the assumptions being made about the endpoints and the values of \(n\). There is a recognition that approximating the integral involves summing the areas of rectangles based on chosen endpoints.

Contextual Notes

Participants note that the problem specifies using left and midpoint Riemann sums, which may impose constraints on how the approximation is approached. There is also mention of the need to define \(n\) for the calculations, which some participants find confusing.

flyingpig
Messages
2,574
Reaction score
1

Homework Statement



Use a left and midpoint Rienman sum to approximate [tex]\int_{0}^{3} x^3 dx[/tex]


The Attempt at a Solution



Left

[tex]\Delta x = \frac{3}{n}[/tex]

Now here is the problem, for left points, what is my 'a'? I can't carve up my intervals and this isn't a right-end approximation so I can't take a= 0 for my xi
 
Physics news on Phys.org
What formula are you using that has an 'a'? Without seeing it, I can't tell you what it is.
 
[tex]x_i = a + i\Delta x[/tex]
 
What is [tex]x_0[/tex] in this case?
 
For clarification, what is the left endpoint of the interval over which you are approximating? That's what your 'a' is supposed to be here. Your function shouldn't affect your choice of a.
 
What do you mean...?
 
a is 0.

x0 = 0 + 0*3/n
x1 = 0 + 1*3/n = 3/n
x2 = 0 + 3*3/n = 6/n
.
.
.
xn = 0 + n*3/n = 3
 
Last edited:
Is the answer for the left end and right end going to be the same for this question?
 
  • #10
But that's what I got...

[tex]\lim_{n\to\infty} \sum_{i=1}^{n} f(x_i)\Delta x[/tex]

With [tex]\Delta x[/tex] the same for all the f(xi)

Like i isn't given I know (I drew a picture) that you don't evaluate f(xi) at the first x1, you do it for x2
 
  • #11
You're getting confused between the partition and the xi values. Suppose you partition the interval [0, 3] into 10 subintervals of equal length. Your partition will have 11 numbers in it:
x0 = 0 + 0 * 3/10 = 0
x1 = 0 + 1 * 3/10 = 3/10
x2 = 0 + 2 * 3/10 = 6/10
x3 = 0 + 3 * 3/10 = 9/10
.
.
.
x10 = 0 + 10 * 3/10 = 30/10 = 3

You can estimate the value of the integral by this sum:
[tex]\sum_{i=1}^{10} f(c_i)\Delta x[/tex]

where ci is any number in the i-th subinterval. Notice that I removed the limit part.

For example, c0 could be the left endpoint (x0), the right endpoint (x1), the midpoint ( (x0 + x1)/2), or any other point in the first interval.

If you take a larger value of n, you'll probably get a closer estimate.
 
  • #12
No, the value of n is not given
 
  • #13
Then you pick a value of n.
 
  • #14
You aren't suppose to, you are suppose to use the summation properties at the end like

[tex]\sum_{i=1}^n i = \frac{n(n+1)}{2}[/tex] to evaluate the sum. I could do it for right-end sums, but I can't do it for left end
 
  • #15
Then leave it at an unknown n. Therefore, every partition has a width of 3/n. Show us your work for the right side. They should be extremely similar - differing only by which side of each partition you plug into f.
 
  • #16
flyingpig said:
Use a left and midpoint Rienman sum to approximate ...
I think you're supposed to pick a value for n to come up with an actual estimate.
 
  • #17
[tex]\lim_{n\to \infty}\sum_{i=1}^{n} f(x_i) \Delta x[/tex]

[tex]\Delta x = \frac{3}{n}[/tex]

[tex]x_i = 0 + i\frac{3}{n}[/tex]

[tex]f(x_i) = \frac{9i^2}{n^2}[/tex]

[tex]\frac{3}{n} \lim_{n\to \infty}\sum_{i=1}^{n} \frac{9i^2}{n^2}[/tex]

Sum this up and it turns out to be exactly as the integral for a right-end approximate
 
  • #18
flyingpig said:
[tex]\lim_{n\to \infty}\sum_{i=1}^{n} f(x_i) \Delta x[/tex]

[tex]\Delta x = \frac{3}{n}[/tex]

[tex]x_i = 0 + i\frac{3}{n}[/tex]

[tex]f(x_i) = \frac{9i^2}{n^2}[/tex]

[tex]\frac{3}{n} \lim_{n\to \infty}\sum_{i=1}^{n} \frac{9i^2}{n^2}[/tex]

Sum this up and it turns out to be exactly as the integral for a right-end approximate
I'm reasonably certain this is NOT what they're looking for. The tipoff is that they have asked you to approximate the integral by a Riemann sum.
 
  • #19
This was an example question for exam. I'll pull one from my midterm and I'll pull out the whole solutions [PLAIN]http://img21.imageshack.us/img21/6295/unledel.png
 
Last edited by a moderator:
  • #20
Compute and approximate are different. You said approximate. Hence Mark's advice.
 
  • #21
Also, I can't help but notice your sum used x^2, while the problem stated x^3
 
  • #22
Oh is it too late to ask how to "compute" for leftend approximate now...?
 
  • #23
For your right hand side, you took the right endpoint of each section in the partition, and plugged it into the formula, correct? For the left hand side, we just plug in the left endpoint:

[tex]\Delta x = \frac{3}{n}[/tex]

[tex]\sum_{i=1}^n \frac{3}{n} f(x_{i-1})[/tex]

That i-1 index is the key here.

What is [tex]x_{i-1}[/tex] in terms of n and i?
 
Last edited:
  • #24
flyingpig said:
Oh is it too late to ask how to "compute" for leftend approximate now...?
Pick a number n.
Divide the interval [0, 3] into that many subintervals.
Evaluate f at the left end of each subinterval.
Multiply by [itex]\Delta x[/itex].
Add the n terms.
 
  • #25
So n has to be defined...? But this doesn't apply to MidPoint right?
 
  • #26
flyingpig said:
So n has to be defined...? But this doesn't apply to MidPoint right?
Yes, n has to be defined. You haven't done this before?

Yes, you need to pick an n to do a midpoint approximation.

Pick a number n.
Divide the interval [0, 3] into that many subintervals.
Evaluate f at the [STRIKE]left end[/STRIKE] midpoint of each subinterval.
Multiply by [itex]\Delta x[/itex].
Add the n terms.
 
  • #27
If your professor meant for you to compute, then by all means let n go to infinity. If he wanted an approximation, do as Mark says.
 
  • #28
Mark44 said:
Yes, n has to be defined. You haven't done this before?

Yes, you need to pick an n to do a midpoint approximation.

Pick a number n.
Divide the interval [0, 3] into that many subintervals.
Evaluate f at the [STRIKE]left end[/STRIKE] midpoint of each subinterval.
Multiply by [itex]\Delta x[/itex].
Add the n terms.

No I have done approximation before, in fact I am really good at them. While everyone (including my TAs) just memorize the formulas, I go and use pictures to help me.
 
  • #29
Sethric said:
If your professor meant for you to compute, then by all means let n go to infinity. If he wanted an approximation, do as Mark says.

But the question said "right Reinman"
 
  • #30
Right Riemann just means to create rectangles based on the height of f(x) on the right hand side of each section in the partition. We really only take the limit of sums when we are dealing with whether something is integrable.
 

Similar threads

  • · Replies 4 ·
Replies
4
Views
2K
Replies
4
Views
3K
  • · Replies 7 ·
Replies
7
Views
2K
  • · Replies 16 ·
Replies
16
Views
2K
  • · Replies 11 ·
Replies
11
Views
3K
  • · Replies 3 ·
Replies
3
Views
2K
  • · Replies 14 ·
Replies
14
Views
2K
  • · Replies 9 ·
Replies
9
Views
2K
  • · Replies 4 ·
Replies
4
Views
3K
  • · Replies 2 ·
Replies
2
Views
2K