Write the regression equation. Y = 83.23009 + 2.29018x1 + 1.30099x2 2.

example from the professor, I need help with part 7, I do not how to find indication of multidisciplinary.

1. Write the regression equation.

Y = 83.23009 + 2.29018x1 + 1.30099x2

2. Interpret the regression constant and partial regression coefficients.

  When newspaper and tv advertising are zero weekly gross revenues will be $82,230.09. For each $1000 change in television advertising weekly gross revenues will change in the same direction by $2,290.18, given that newspaper advertising stays constant. For each $1000 change in newspaper advertising weekly gross revenues will change in the same direction by $1,300.99, given television advertising remains constant.

3. Plug in any value for x1 between 2.0 and 5.0 and for x2 any value between 1.5 and 4.2.. For example  Y = 83.23009 + 2.29018 (2.0) + 1.30099 (3.0) = 91.71342 = $91,713.42

4. Test the significance of the partial regression coefficients at alpha = .05.

HO: B1 = 0 Ha: B1 =/ 0 table t value 2.571 Calculated value t = 7.531899 Reject Ho.

HO: B2 = 0 Ha: B2 =/ 0      same                                       t = 4.056697 Reject Ho.

5. Test the overall significance of the regression model.

Ho Not Significant Ha: Significant   table F value 5.79 Calculated value F = 28.37777

Reject HO.

6 Interpret the coefficient of determination. 88.67 % of the variation in weekly gross revenue can be explained by variation in television and newspaper advertising.

7. Are there any indications of multicollinearity?

 The size of the partial regression coefficients are reasonable not too large or small.

 The correlation coefficient between television and newspaper advertising is -.56 well below .8.

 The sign between newspaper advertising and weekly gross revenues in the correlation matrix is negative but in the multiple regression model its positive. As the signs should match there is multicollinearity.