Prove the sample variance formula.

  • Thread starter Thread starter The_Iceflash
  • Start date Start date
  • Tags Tags
    Formula Variance
Click For Summary

Homework Help Overview

The discussion revolves around proving the formula for sample variance, specifically the expression for S² involving sums of squares and the mean of a sample. The subject area is statistics, focusing on variance calculations.

Discussion Character

  • Exploratory, Mathematical reasoning, Assumption checking

Approaches and Questions Raised

  • Participants explore the relationship between sums of squares and the mean, attempting to manipulate the expression for variance. There are questions about the role of the sample mean in summations and how to simplify the expressions correctly.

Discussion Status

Participants are actively engaging with the problem, offering hints and exploring different algebraic manipulations. There is a focus on understanding the implications of treating the sample mean as a constant in summations, and some participants are questioning their own reasoning and assumptions.

Contextual Notes

There are hints provided regarding the treatment of the sample mean and its relationship to the sums involved, indicating that participants are navigating through the algebraic complexities without a clear resolution yet.

The_Iceflash
Messages
50
Reaction score
0

Homework Statement


Prove that the sample variance of a sample is given by
S2 = \frac{\sum^{n}_{j=1}x_j^2 - \frac{1}{n} (\sum^{n}_{j=1}x_j)^2}{n-1}

Homework Equations



N/A

The Attempt at a Solution



For my purposes it is sufficient to show that:

\sum^{n}_{j=1}(x_j -x_j^2) = \sum^{n}_{j=1}x_j^2 - \frac{1}{n}\left(\sum^{n}_{j=1}x_j\right)^2



I got as far as this:

= \sum^{n}_{j=1}x_j^2 - \sum^{n}_{j=1}2\bar{x}x_j + \sum^{n}_{j=1}\bar{x}^2

I need help getting from there to here:

= \sum^{n}_{j=1}x_j^2 - \frac{1}{n}\left(\sum^{n}_{j=1}x_j\right)^2

Thanks in advance and I apologize for any coding error.
 
Last edited:
Physics news on Phys.org
Hint: remember that \overline x is a constant and can be removed from summations.
 
statdad said:
Hint: remember that \overline x is a constant and can be removed from summations.

So if I put the \overline x's in front

= \sum^{n}_{j=1}x_j^2 - 2\bar{x}\sum^{n}_{j=1}x_j + \bar{x}^2\sum^{n}_{j=1}

and rearrange it I get

= \sum^{n}_{j=1}x_j^2 +\bar{x}^2\sum^{n}_{j=1} - 2\bar{x}\sum^{n}_{j=1}x_j

My instincts tell me to factor out the \overline x but I don't see how that could help.
 
The_Iceflash said:
So if I put the \overline x's in front

= \sum^{n}_{j=1}x_j^2 - 2\bar{x}\sum^{n}_{j=1}x_j + \bar{x}^2\sum^{n}_{j=1}

and rearrange it I get

= \sum^{n}_{j=1}x_j^2 +\bar{x}^2\sum^{n}_{j=1} - 2\bar{x}\sum^{n}_{j=1}x_j

My instincts tell me to factor out the \overline x but I don't see how that could help.

Hint:

<br /> \sum_{i=1}^n \overline x_i = \overline x_i \sum_{i=1}^N 1<br />

(you forgot to include the 1: what does the above sum equal?

Hint:

<br /> \overline x = \frac 1 n \sum_{i=1}^n x_i<br />

so what must

<br /> \sum_{i=1}^n x_i<br />

equal?
 
statdad said:
Hint:

<br /> \sum_{i=1}^n \overline x_i = \overline x_i \sum_{i=1}^N 1<br />

(you forgot to include the 1: what does the above sum equal?

Hint:

<br /> \overline x = \frac 1 n \sum_{i=1}^n x_i<br />

so what must

<br /> \sum_{i=1}^n x_i<br />

equal?

So, \sum_{i=1}^n x_i must be equal to n\overline x correct?

The sum from the first sum is n, correct?
 

Similar threads

  • · Replies 6 ·
Replies
6
Views
3K
Replies
3
Views
2K
Replies
0
Views
991
  • · Replies 4 ·
Replies
4
Views
1K
  • · Replies 1 ·
Replies
1
Views
2K
  • · Replies 7 ·
Replies
7
Views
2K
Replies
8
Views
2K
  • · Replies 16 ·
Replies
16
Views
2K
Replies
4
Views
2K
Replies
5
Views
2K