In: Statistics and Probability
Problem 4
The following data exists for Lawrence Repair, a large equipment repair company. The behavior pattern of the maintenance overhead costs must be determined to prepare the annual profit plan for next year. The cost accountant has suggested using statistical analysis to derive an equation in the form of y=mx+b for maintenance overhead costs. Monthly data regarding past parts costs, repair hours and overhead costs are provided below.
Month |
Overhead Costs |
Parts Cost |
Repair Hours |
1 |
9896 |
1068 |
250 |
2 |
9249 |
1455 |
250 |
3 |
13205 |
3503 |
482 |
4 |
10560 |
1371 |
286 |
5 |
9059 |
1547 |
202 |
6 |
10667 |
1225 |
382 |
7 |
12888 |
2989 |
570 |
8 |
10350 |
1844 |
346 |
9 |
11222 |
1657 |
450 |
10 |
13274 |
2103 |
546 |
11 |
10835 |
1248 |
342 |
12 |
12612 |
2703 |
414 |
13 |
10876 |
2203 |
386 |
14 |
12821 |
3113 |
406 |
15 |
8469 |
755 |
214 |
Prepare a graphical representation of overhead cost estimation using Parts Cost as the independent variable. Using Repair Hours as the independent variable. For both scatterplots, use good formatting to include a Figure Title, Axis labels, trend line and remove blank area on the left half of the chart at a minimum.
Compute and report the simple regression results for each of the independent variables.
Report the following from your two outputs |
Parts Cost |
Repair Hours |
Cost Equation using regression output |
||
Predict overhead costs for $3500 parts cost & 480 repair hours |
||
How much of the change in overhead can be explained? |
||
Indicate the t-statistics, is is good? |
||
Indicate the p-value, is is good? |
||
Construct a 95% confidence interval for the estimated coefficient. |
||
Which activity (parts cost or repair hours) do you believe best explains the variation in overhead costs? Provide your justification from the chart. |
Compute and report a multiple regression analysis using both independent variables to explain the variation in costs. Provide a single cost equation using both independent variables.
Based on the single cost equation derived from the regression analysis, predict overhead costs for $3500 parts cost & 480 repair hours.
What is the percent of the total variance that can be explained by the regression equation?