- #1
MaxManus
- 277
- 1
Homework Statement
Given a simple linear model y = B1 + B2*x + e and the least square estimators, we can estimate E(y) for any value of x = x0 as Y0 = b1 + b2*x0
Describe the difference between predicting y0 and estimating E(y0)
Homework Equations
The Attempt at a Solution
I am not sure what the difference is.
Prediction:
Y= b1 + b2*x0
f = y0 - Y0 = (B1 +B2*x0 + e0) - (b1 + b2*x0)
E(f) = 0
and var(f) = sigma2*(1 + 1/N + (x0 - [tex]\bar{x}[/tex])2/sum(xi - [tex]\bar{}x[/tex])2
where [tex]\bar{}x[/tex] is the average.
the prediction interval is Y0 +- tc*se(f)
where se is the standard error of the forcast.
Can anyone help me with estimating E(y0)?