Solved

# Multiple linear regression

Posted on 2008-10-15
Medium Priority
359 Views
My science calculator can calculate the MSE value of a Quadraic equation only. But I need to find out the S(d0)(=Root(S(y0)*S(y0)+MSE). If I get the value of MSE, in normal, what value will be the S(y0)*S(y0)? This means if the value of S(y0)*S(y0) is normally 1/3 or 1/4 (or other number) of MSE, then, I can estimate my S(d0).
0
Question by:w0168

LVL 9

Expert Comment

ID: 22721740
It's not really clear what you're asking from all the notation.  I assume by MSE, you're referring to mean square error?  MSE is a very general term...  but the title of your question is multiple linear regression.  So are you referring to a regression concept (like the standard error of the estimate?)  S(y0) could refer to the standard deviation of y or predicted y, and S(y0)*S(y0) the variance, but it's unclear from what you're asking.

Could you clarify your notation?  And perhaps state the purpose behind this question?  That will help in answering it.  In what context do you need an answer?  Is this being used in a web app somehow?
0

Accepted Solution

w0168 earned 0 total points
ID: 22727188
Yes, I need to know the standard deviation,S(d0), of the predicted y (for a Quadratic equation). My calculator can find out its mean square error (MSE), but no variance covariance matrix function to provide the required values,like S(b0),S(b0,b1),S(b0,b2),S(b1),S(b1,b2),S(b2), for calculating S(y0). Because I need a rough estimation only, so, if an experienced expert may tell me how to estimate S(y0) value if I have only MSE value.
0

## Featured Post

Question has a verified solution.

Are you are experiencing a similar issue? Get a personalized answer when you ask a related question.

Have a better answer? Share it in a comment.