The Math Forum

Search All of the Math Forum:

Views expressed in these public forums are not endorsed by NCTM or The Math Forum.

Math Forum » Discussions » sci.math.* » sci.stat.math

Notice: We are no longer accepting new posts, but the forums will continue to be readable.

Topic: comparison between two value in a sample, please help...
Replies: 3   Last Post: Nov 18, 2012 10:02 AM

Advanced Search

Back to Topic List Back to Topic List Jump to Tree View Jump to Tree View   Messages: [ Previous | Next ]
Jinsong Zhao

Posts: 22
Registered: 11/27/10
Re: comparison between two value in a sample, please help...
Posted: Nov 18, 2012 10:02 AM
  Click to see the message monospaced in plain text Plain Text   Click to reply to this topic Reply

Thank you very much for your kindly help, and sorry for no feedback for
such long time. I have some questions about this problem. If possible,
please give me help. Thanks in advance.

In the following, I will do the calculation according to your comments.
Please tell me whether I am wrong or not.

On 2012-08-18 9:39, Rich Ulrich wrote:
> On Fri, 17 Aug 2012 01:49:39 -0400, Rich Ulrich
> <> wrote:
> [snip, a bunch]

>> I *think* what follows is a safe, conservative procedure --
>> The minimum, proper fitted equation will show no less error
>> than the measurement error. For that, the pooled SD is

> Keep in mind that the MS-residual is an estimate of the variance
> (which is the square of the standard deviation) of the residual.
> When I was learning to use regression, it helped me when
> I realized that fact. Then I started making a point of noticing that
> variance, and taking the square root ... just to make sure that
> I was still analyzing the same set that I thought I was, and that
> nothing was going strange.

>> about 42, or the "Mean-square-residual" is the square of
>> 42, with 10 d.f. Take the Mean square for Regression from
>> the 5-point regression and divide it by *this* residual,
>> instead of the computed residual, to get the safe F-test.

> Oops! slligh correction -- For the 5 point regression, each
> point is the average of three measures. The SD of the *3*
> taken separately is 42; the SD of the average is, naturally,
> the raw SD divided by the square root of N, or sqrt(3).
> Or - the MS is not 42-squared, but is 42-squared, divided
> by 3.

Pooled SD:

sqrt(sum((33.69357 27.84428 33.20421 57.29368 51.11501)^2)*2/10) = 42.20993

i.e., Mean-square-residual = 42.20993^2/3 = 593.8927

The Analysis of Variance Table for the 5-point regression is:

Response: y
Df Sum Sq Mean Sq F value Pr(>F)
x 1 39777 39777 15.48 0.02924 *
Residuals 3 7709 2570

So, the Mean square for Regression is 39777. Then the safe F-test is:

F_value = 39777/593.8927 = 66.97675 on 1 and 3 DF.

The corresponding P value is 0.003816985.

My question is what's the meaning of this P value? Does it mean that the
regression is significant? If so, does it mean that the value of N in
1952 differs with that in 1998?

If the P value large that 0.05 (and alpha is set to 0.05), then how to
interpret the result?

What I hope to know is whether the value of N in two different year is

>> The ratio of the two so-called Residual terms can be used as
>> a test of whether the regression is "over fitted".

Would you like to tell me how to form the ratio? 2570/593.827 on 3 and 3 DF?

> Still useful, but you do use the corrected value.

Thanks again for your kindly help.


Point your RSS reader here for a feed of the latest messages in this topic.

[Privacy Policy] [Terms of Use]

© The Math Forum at NCTM 1994-2018. All Rights Reserved.