The Coefficient of Determination is the proportion of the explained variation divided by the total variation, when a linear regression is performed.

The Coefficient of Determination is R^2. The square of the linear coefficient is r^2.

R^2 = r^2 = Sxy^2 / Sx2 * Sy2

0 <= r^2 <= 1

- R^2 will = +1 only when ALL of the points fall exactly on the line. (SSE (the sum of squared errors) will = 0)
- As the scatter points around the line become greater, the SSE term will become greater, thus r^2 will go down.
- As the scatter points around the line become less, SSE becomes less, thus r^2 will approach 1.

## Leave a Reply