# Thread: Standard error of estimate

1. ## Standard error of estimate

I had a hw assignment in my statistics class where I got the answer for the problem, but now I need to do the standard error of estimate and have no idea how to do it, and I do not understand the books or teachers explanation.

For part A the question was assuming a linear relationship, derive the least squares regression line for predicting self esteem from religious involvement.

The answer I got was :y=-0.075x+6.489

For part B the question was Using this regression line, what value of self esteem would you predict for an eight grader who had value of religious involvement?

So i need help finding the standared error of estimate for this problem which is a whole

The answer for part B is 3.28

2. Originally Posted by Girlaaaaaaaa
I had a hw assignment in my statistics class where I got the answer for the problem, but now I need to do the standard error of estimate and have no idea how to do it, and I do not understand the books or teachers explanation.

For part A the question was assuming a linear relationship, derive the least squares regression line for predicting self esteem from religious involvement.

The answer I got was :y=-0.075x+6.489

For part B the question was Using this regression line, what value of self esteem would you predict for an eight grader who had value of religious involvement?

So i need help finding the standared error of estimate for this problem which is a whole

The answer for part B is 3.28
1. You do not state what the value of the religious involvement variables is for part B.

2. Part B does not require a standard error estimate as posted.

CB

3. Those are the correct answers(were in back of the book even) and with those I need to find the Standard error of estimate. Which is what I am asking help with for if you re-read the post

4. Originally Posted by Girlaaaaaaaa
Those are the correct answers(were in back of the book even) and with those I need to find the Standard error of estimate. Which is what I am asking help with for if you re-read the post
You last sentence starts "So" which seems to imply that this is a consequence of what went before, but it does not.

Also the given answer to part B. cannot be obtained from what you post.

If you have no additional information and the underlying assumptions for linear regression hold I would use the standard deviation of the residuals for the SE of a prediction (the residuals should be iid normal random variables, and the mean reasidual should be zero about a regression line).

CB

5. If you read the beginning of my post it says "I had a hw assignment in my statistics class where I got the answer for the problem, but now I need to do the standard error of estimate and have no idea how to do it, and I do not understand the books or teachers explanation." So yes there was more information, i just posted the answers I got though. So now I need to do the standard error of estimate, which is what I am not sure how to do.