Carpet Company's ABC analysis indicates that Manufacturing overhead costs include large costs incurred for setups when a new batch of carpet is started. Carpet Company's ABC analysis indicates that Manufacturing overhead costs include large costs incurred for setups when a new batch of carpet is started. Carpet Company's ABC analysis indicates that Manufacturing overhead costs include large costs incurred for setups when a new batch of carpet is started.
Week |
Number of Batches |
Direct Manufacturing Labor-Hours |
Manufacturing Overhead Cost |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
12 |
30 |
$ 1,190 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2 |
15 |
72 |
1,211 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
3 |
13 |
36 |
1,004 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
4 |
11 |
20 |
917 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
5 |
12 |
45 |
1,456 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
6 |
16 |
44 |
1,180 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
7 |
7 |
38 |
710 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
8 |
14 |
70 |
1,316 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
9 |
12 |
30 |
1,032 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
10 |
7 |
29 |
752 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
11 |
8 |
33 |
800 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
12 |
10 |
35 |
865 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
Do not retype the information above. You should use this sheet as your data sheet and your solution should be on separate |
|
|
|
|
|
|
|
|
|
|
|
sheets. |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
Part 2 should contain 2-simple regression tables, 1-multiple regression table, and 2-scattergraphs when you are done.--I have done one of the scattergraphs and simple regression using direct labor for you. |
|
|
Required: |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1. Prepare 2 scattergraphs and 2 simple regression analyses***** to estimate the Manufacturing Overhead costs using each of the |
|
|
|
|
|
|
|
|
|
|
cost drivers above. Make sure you show the equation of the line and R-squared on each of your scattergraphs. |
|
|
|
|
|
|
|
|
|
|
|
|
I have gone ahead and developed the simple regression table and the scattergraph using Direct Manufacturing labor hours as the cost driver (I.e. independent variable) |
|
|
|
|
|
You do not need to redo these, but you may want to use these as a verification that you know how to develop a scattergraph and a simple regression table. |
|
|
|
|
|
|
|
The output and chart for your simple regression should be on separate sheets that should be named appropriately. |
|
|
|
|
|
|
|
|
|
|
|
|
Type the equation of the line under the regression output on each of the simple regression worksheets. |
|
|
|
|
|
|
|
|
|
|
|
|
|
To make a scattergraph with data from nonadjacent columns highlight the X data (i.e. independent variable-cost driver) first and then hold down the |
|
|
|
|
|
|
|
|
|
control (ctrl) key and highlight the Y data (i.e. the dependent variable-cost). |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
If you have done your scattergraphs correctly the equation of the line and the R-squared should match the output you have on your regression table. |
|
|
|
|
|
|
|
|
|
(i.e. R-squared, Intercept, and X Variable.) Hint: the Y(dependent) variable is the same in all your scattergraphs and regression analyses--which is Manufacturing overhead cost. |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2. Prepare a multiple regression table using batches and direct labor hours for the independent variables. Do not try to prepare a scattergraph, since you cannot make a chart with multiple X's. |
|
|
|
|
|
Using the output from the multiple regression, give the equation to estimate manufacturing overhead costs in terms of these |
|
|
|
|
|
|
|
|
|
|
|
2 cost drivers. Type out the equation of the line on your multiple regression worksheet. |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
X1 |
Number of batches |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
X2 |
Direct Manufactuirng labor hours |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
3. Assuming the following level of cost-driver volume, what is the estimated Manufacturing overhead |
|
|
|
|
|
|
|
|
|
|
|
|
|
cost using simple regression for each of the cost drivers and using multiple regression. You should have |
|
|
|
|
|
|
|
|
|
|
|
|
|
3 computations for this answer. I have completed one of the answers for you. You must use cell referencing from your regression tables to compute the estimated cost to earn credit for part 3. |
|
|
|
|
Cost Drivers (independent variables): |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
Number of batches |
13 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
Direct Manufacturing labor hours |
37 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
Solution: |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
Direct Manufacturing labor hours |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
Answer: |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
$ 1,008.78 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
You need to answer: |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
Number of batches |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
Answer: |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
Multiple regression |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
Answer |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
4. Using the simple and multiple regression analyses output tables, what independent variable(s) would you recommend |
|
|
|
|
|
|
|
|
|
|
|
to estimate the Indirect Manufacturing Overhead Costs? Why? Your response should be at least 75 words and discuss the R-Squared. |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|