|
|
|
Next: Inferences on a prediction
Previous: Inferences on the estimated slope
Original Model: Yi = b0 + b1Xi + ei,where ei ~ N(0,s2) and cov(ei,ej)=0 for i¹ j.
Yh=b0+b1xh (the fitted value for a given xh)
,
))
is estimated by using the sample variance s2 in place
of s2 above and is denoted by 
.
Thus t = 
follows a t distribution with (n-2)
d.f.
Hence a 100(1-a )% confidence interval
for (
)
is given by
|
|
|
Next: Inferences on a prediction
Previous: Inferences on the estimated slope
