Variance of a weighted population

2 replies · 2K views · last reply
rogo0034
Messages
36
Reaction score
0

Homework Statement


Textbook problem 8.23 about a random variable X for cherries in a cherry puff, with a table of x = 4, 5, 6, 7 and P(X = x) = 0.2, 0.4, 0.3, 0.1, followed by parts (a) to (c) on mean, variance, and a probability for the sample mean of 36 puffs.



Homework Equations





The Attempt at a Solution



Was able to get the population mean of 5.3 with 4*.2 + 5*.4 + 6*.3 + 7*.1
But for some reason I'm drawing blank for the variance (which i guess is .81)
 
rogo0034 said:

Homework Statement


Problem 8.23 text with a table of probability distribution for random variable X over values 4, 5, 6, 7 with probabilities 0.2, 0.4, 0.3, 0.1, and parts (a) to (c) asking for mean, variance, and a probability about cherries per puff.

Homework Equations


The Attempt at a Solution



Was able to get the population mean of 5.3 with 4*.2 + 5*.4 + 6*.3 + 7*.1
But for some reason I'm drawing blank for the variance (which i guess is .81)

Variance is defined as:
[tex]var\{x\}=E\{(x-E\{x\})^2\}[/tex]
So if you wanted to directly apply the equation above, you would compute it the same way as mean, except instead of doing a weighted sum of x, you would do a weighted sum of (x-E{x})^2. So for each x, before doing the weighted sum, you would subtract off the mean you found already and square it.

But there is a popular reworking of the equation:
[tex]var\{x\}=E\{x^2-2xE\{x\}+E^2\{x\}\}=E\{x^2\}-E^2\{x\}[/tex]

So in this case, you would compute the expected value of x^2 (the same way you did the mean except use x^2 instead of x). Then, subtract the square of the expectation you already found.