How to Manage Your Belly Fat with Good Nutrition and Exercise

Introduction 

The Good Belly management asked their marketing manager to justify the marketing expenses incurred during promotions of the products. The marketing manager decided to ask one of his employees to use a statistical approach to justify the budget. This was specifically through determining the impact of the marketing activities on the weekly sales. This research considered nine variables, where one (weekly sales) was the dependent variable and the other eight were the independent/ response variables. A regression analysis approach was used for analysis. At first, all the variables were included in the model and the insignificant variables dropped. The second step contained only the significant variables in the model. 

Step 1: All Variables Included 

The tables below show results from the multiple regression analysis. 

From the regression statistics above, the independent variables have a correlation of 82.01% with the dependent variable. Additionally, the R- square value revealed that the independent variables explained 67.26% of the total variation in weekly sales. 

The table above shows the analysis of variable results, where weekly sales was the dependent variable and the other eight variables the independent variables. From the result, the model is significant in explaining the relationship between weekly sales and the independent variables, F = 353.66, p < 0.000. 

The table above shows the model coefficients for the model variables. From the results, the first six variables are significant predictors of weekly sales, since they have a p-value of less than 0.05. The other two variables were not significant predictors. These variables were dropped from the model to improve it. 

Since this model contains insignificant variables, it required improvements in order to predict the weekly sales significantly. The section below shows regression analysis by considering only the significant independent variables. 

Step 2: Significant Variables Included 

The section below shows the regression analysis with the six significant independent variables. 

The regression statistics above show that the independent variables and weekly sales have a correlation of 81.99%. Additionally, the r-square value shows that the independent variables explain 67.23% of the total variation in weekly sales. 

The analysis of variance above revealed that the model with six significant independent variables was significant in showing the relationship between the variables and weekly sales, F = 471.4447, p < 0.000. 

The table above shows the model summary results. From the results, all the six variables are significant predictors of weekly sales, since they have p-values less than 0.05. Additionally, the results showed that a unit increased in average retail price led to a decrease of weekly sales by approximately -28.61 units. A unit increase in sales representatives, increased the weekly sales by 76.95 units. A unit increase in endcap promotions increased the weekly sales by 304.96 units. A unit increase in demos in the corresponding week increased the sales by 111.26 units. A unit increase in demos 1 to 3 weeks ago increased the weekly sales by 73.66 units. Lastly, an increase in demos 4 to 5 weeks ago led to an increase in weekly sales by 67.70 units. 

From the results above, the best regression model to explain changes in the weekly sales is as given below: 

Weekly sales = 294.19 – 28.61 (Average retail price) + 76.95 (Sales rep) + 304.96 (Endcap) + 111.26 (Demo) + 73.66 (Demo 1-3) + 67.70 (Demo 4-5) 

Conclusions 

The management of Good Belly sought to understand the most effective marketing strategies using a statistical approach. This was achieved by checking the significant predictors of weekly sales. From the analysis above, average retail price, sales representatives, endcap, demos in corresponding weeks, demos in the previous 1-3 weeks and demos in the previous 4-5 weeks were significant predictors of weekly sales.