isi

[ritu]

Consider the two variable linear regressn model where error Ei are iid with mean 0 nd  variance 1..
Yi=A+B(Xi-X bar)+Ei
i is 1,2,3,,,,,,,,,
let a nd b  be ols estimates of A nd B..find correlatn coefficient btwn a nd b...
Isi 2010

[Amit Goyal]

0

[ritu]

Why not -1 sir?

[ritu]

Is it because from given regressn equatn we can derive the result that Ybar =a..that means a is independent of b nd hence uncorrelated???

[anon_econ]

Ritu, my teacher in college told me that there's a formula for the coefficient of correlation b/w the least squares estimators..u can try looking for it. i don't know if there's another method to find it.

[anon_econ]

ok the formula is given on page 74 here:
http://www.stat.lsa.umich.edu/~kshedden/Courses/Stat600/Notes/least-squares.pdf
it'll give u zero

[Amit Goyal]

For Proof: Refer problem number 15 on pages 7 - 8 of the following http://economicsentrance.weebly.com/uploads/1/1/0/5/1105777/notes_1.pdf