Please help.

In a multiple linear regression model, y = Bo + Bix, + 82x2 + &, where x1 and x2 are non-random independent variables, and & is the random error. Now a set of n items give data, ()1, X11, X21), ..., (Vn, Xin, X2n) , which follow this model. Decompose total sum of squares into a set of sums of squares for testing statistical significance of the regression model and for testing statistical significance of each regressor, respectively, under Scenario 1: Bo = 0 Scenario 2: Bo # 0 Scenario 3: Bo may or may not be zero