Regression Model of Predicting the Advertising Sales

Regression Model A

To predict some advertising sales based on the number of viewers and the length of the program, a regression model was used, and the following results were obtained:

Y = bx1 + bx2 + c

Advertising sales = No. of views (x1) + Length (x2) + c

Advertising sales = 796.7551062x1 - 11.54719579x2 + 16583.33382

Based on the regression model of predicting the advertising sales using the number of viewers and the length of the program, it can be deduced that advertising sales have a direct relationship with the number of viewers and an indirect relationship with the duration of the program. Thus, an increase in the number of viewers by 1 million leads to the growth in the advertising sales by $796.76 while an increase in the length of the program by one minute leads to decrease in advertising sales by $11.547. Advertising sales amount to $16,583.3 when the no views or no length of the program. The number of viewers and duration of the program both explain about 67.53% of the variation in the advertising sales data or 62.12 percent when adjusted to the degrees of freedom. The data set is highly correlated as their person correlation coefficient is equal to 0.8218 which is closer to 1. Thus, the forecast of the special aimed at teen viewers can

Since the P-value (0.000338) is less than the significance value (0.05), we conclude that the results are significant and that the special would bring in 3.2 million viewers.

Regression Model B

To predict some advertising sales based on the number of viewers, the length of the program, and the average viewer age, a regression model was used, and the following results were obtained:

Y = bx1 + bx2 + c

Advertising sales = No. of views (x1) + Length (x2) + Average Viewer Age (x3) + c

Advertising sales = 756.15108x1 -24.56121x2 + 95.442684x3 + 15041.564

Advertising sales can be seen to have a direct relationship with the number of views and average viewer age and an indirect relationship with the length of the program. In this case, an increase in the number of viewers by 1 million leads to the growth in the advertising sales by $756.15 while the growth in the average viewer age for one year leads to an increase in the advertising sales by $95.44. On the other hand, an increase in the length of the program by 1 minute leads a decrease in the advertising sales by $24.56. The number of viewers, the term of the program and the average viewer age jointly explain about 67.53 percent of the variation in the advertising sales data or 64.10 percent when adjusted to the degrees of freedom. The correlation coefficient is 0.8473 implying that the data set exhibits a high correlation between each other.

Since the P-value (0.000621) is less than the confidence level of 0.05, we conclude that statistics were significant and that the special would bring in 14.1 viewers and the average viewer age would be 32.