Size: 1289
Comment: Rewrite 1
|
Size: 1253
Comment: Rewrite 2
|
Deletions are marked like this. | Additions are marked like this. |
Line 11: | Line 11: |
As a starting point, the regression line must pass through these two points: | This line must pass through the mean and the slope of the line must be the marginal change in ''Y'' given a unit change in ''X''. In other words, the line must pass through two points: |
Line 13: | Line 13: |
{{attachment:regression1.svg}} | {{attachment:model3.svg}} |
Line 15: | Line 15: |
and | Insert the first point into the estimation. This is quickly solved for ''α''. |
Line 17: | Line 17: |
{{attachment:regression2.svg}} | {{attachment:alpha1.svg}} |
Line 19: | Line 19: |
Take the generic equation form of a line: {{attachment:b01.svg}} Insert the first point into this form. {{attachment:b02.svg}} This can be trivially rewritten to solve for ''a'' in terms of ''b'': {{attachment:b03.svg}} |
{{attachment:alpha2.svg}} |
Ordinary Least Squares Univariate Proof
The model is constructed like:
This is estimated as:
This line must pass through the mean and the slope of the line must be the marginal change in Y given a unit change in X. In other words, the line must pass through two points:
Insert the first point into the estimation. This is quickly solved for α.
Insert the second point into the original form.
Now additionally insert the solution for a in terms of b.
Expand all terms to produce:
This can now be eliminated into:
Giving a solution for b:
This solution is trivially rewritten as:
Expand the formula for correlation as:
This can now be eliminated into:
Finally, b can be eloquently written as:
Giving a generic formula for the regression line: