Regression Analysis of Advertising Sales

Based on the multiple regression analysis output in the excel sheet “Model A”, the regression model is estimated to be;

Advert. Sales= 16583.3382 + 796.75510621 Av. Viewers - 11.54719579 Prog. Length.

That is,

Y = 16583.33382+796.75510621 X1-11.54719579X2

The regression analysis was conducted at 5% level of significance. From the regression outputs, R2= 0.675293897, implying that 67% of the variations in advertisement sales are caused by the explanatory variables (average viewers and program length). Using the p-value, we examine the significance of the regression coefficients at 5% level of significance. The autonomous sales and coefficient for average viewers are statistically significant at alpha= 0.05 since their p-values are less than 0.05. However, the regression coefficient for program length is not statistically significant at 5% level of significance since its p-value is higher than 0.05 (Draper & Smith, 2014).

The amount of advertisement sales can be estimated by replacing the forecasted values in the equation

Y = 16583.33382+796.75510621 X1-11.54719579X2

Y = 16583.33382 +796.75510621 (3200000)-11.54719579(60)

ð $ 2,549,632,230.33

Model B

From the regression analysis output for the model, the following model estimation is obtained;

Advert. Sales = 15041.5642+756.1510757Av. Viewers - 24.56120666 Prog. Length + 95.44268428 Av. Age

That is;

Y= 15041.5642+756.1510757X1-24.56120666X2+95.44268428X3

Since the multiple regression was conducted at 5% level of significance, it can be concluded that 71% of the variations in advertisement sales are caused by the three explanatory variables since R= 0.717941637. The p-values for the regression of autonomous sales and average viewers are statistically significant at 5% level of significance since their p-values are greater than 0.05 (Draper & Smith, 2014). The p-values of the regression coefficients of the variables program length and average age of viewers are higher than 0.05 implying that they are not statistically significant.

From the model, the amount of advertisement sales can be estimated by replacing the forecasted values of the three variables in the equation;

Y= 15041.5642+756.1510757X1-24.56120666X2+95.44268428X3

Y= 15041.5642+756.1510757 (120)-24.56120666(14100000)+95.44268428(32)

ð $ 10,661,745,315.06

Reference

Draper, N. R., & Smith, H. (2014). Applied regression analysis(Vol. 326). John Wiley & Sons.