Question

Lo. Look at the following data table indicating the actual # of units sold as well as 3 other factors during the 5 years the company produces volleyballs.

units sold	price	Amount of rainy Days per year	Advertising dollars
190	$20.00	131	$50,000.00
172	$22.00	173	$35,000.00
231	$25.00	82	$70,000.00
266	$28.00	67	$70,000.00
192	$31.50	123	$55,000.00

1. ) Using the data above
   1. Provide the R-squared value for each independent variable, using Units Sold as your dependent variable.
   2. What variable is the best predictor of Units Sold?
   3. What variable is the worst predictor of Units Sold?
   4. Provide 2 actions you would take with this information?

2. (4 Marks) Compare the actual units sold table with the initial prediction of the Unit Sold growth of 10% per year.
   1. What is the total MAE across the 5 years?
   2. What is the total MAPE across the 5 years?
   3. Was their initial forecast and forecasted growth accurate or inaccurate, why?

Answer 1

Solution:

Please find the summary output for all the three variables below,

a)

SUMMARY OUTPUT Price Regression Statistics Multiple R 0.328523 R Square 0.107928 Adjusted R -0.18943 Standard E 41.33109 Observatic

SUMMARY OUTPUT Rainy Days Regression Statistics Multiple R 0.948729 R Square 0.900086 Adjusted R 0.866781 Standard E 13.83215 Observatic

SUMMARY OUTPUT Advertising Amount Regression Statistics Multiple R 0.900871 R Square 0.811568 Adjusted R 0.748757 Standard E 18.99566 Observatic

Hence, The R-Squared values are the ones that are highlighted in Yellow.

b)
The best predictor for the number of units sold is Rainy days because the R-Squared value is highest meaning the rainy day's variables best explain the variation in the units sold. This is why the best predictor of the number of units sold is rainy days because of its high R-Squared value which determines the model fit.

c)
The worst predictor is the price because it least explains the variation in the number of units sold and hence is the worst predictor of the number of units sold.

d)
With the given information we can determine the prediction model that will help us in predicting future sales for the company. This gives us an approximate value with which we can work. This approximate value is known as the predicted value which is predicted by the model that is based on past sales data. This is why we will be in a better position to predict future sales and be operationally ready to meet consumer demand.

Secondly, we can understand the factors on which the demand is dependent. Here, we came to know that the price is the worst indicator for the future sales predictor hence we can be in confidence that the increase in prices will not affect the demand much. Hence, such information has a big business advantage.

**Can solve only first four parts as per HOMEWORKLIB POLICY**

We were unable to transcribe this image

SUMMARY OUTPUT Rainy Days Regression Statistics Multiple R 0.948729 R Square 0.900086 Adjusted R 0.866781 Standard E 13.83215 Observatic

SUMMARY OUTPUT Advertising Amount Regression Statistics Multiple R 0.900871 R Square 0.811568 Adjusted R 0.748757 Standard E 18.99566 Observatic