Running Head: COST ESTIMATION USING REGRESSION ANALYSIS

COST ESTIMATION USING REGRESSION ANALYSIS

Name of Student
Name of Course
Course Instructor
Date
 COST ESTIMATION USING REGRESSION ANALYSIS

Question 1

In order to determine which cost driver is better among direct labor cost and direct labor
hours for estimating direct labor related manufacturing overheads, we have generated two
different regression models for Option A and Option B. The results of Option A are as follows:

Regression Statistics
Multiple R 0.63
R Square 0.39
Adjusted R
Square 0.36
Standard
Error 2431.72
Observations 20

ANOVA
Significan
df SS MS F ce F
68894461.8 68894461
Regression 1 3 .83 11.65 0.00
106438927. 5913273.
Residual 18 81 77
175333389.
Total 19 64

Coefficie Standard P- Lower Upper Lower Upper

nts Error t Stat value 95% 95% 95.0% 95.0%
29196.1
Intercept 23260.16 2825.44 8.23 0.00 17324.13 9 17324.13 29196.19
Direct Labor
cost -0.14 0.04 -3.41 0.00 -0.23 -0.06 -0.23 -0.06
 Direct Labor cost Re-
sidual Plot
10000
5000

Residuals
0
-5000 00 00 00 00 00 00 00
0 ,0 0 ,0 0 ,0 0 ,0 0 ,0 0 ,0 0 ,0
4 5 6 7 8 9 10
Direct Labor cost

Normal Probability Plot

Direct Labor-related MOH

25000
20000
15000
10000 Series1
5000
0
5 5 5 5 5 5 5 5 5 5
2. 12. 22. 32. 42. 52. 62. 72. 82. 92.
Sample Percentile

The results of option B for direct labor hours are shown below:

Regression Statistics
Multiple R 0.99
R Square 0.98
Adjusted R
Square 0.98
Standard
Error 479.11
Observations 20

ANOVA
Significan
df SS MS F ce F
171201637. 17120163 745.8
Regression 1 48 7.48 4 0.00
Residual 18 4131752.16 229541.79
175333389.
Total 19 64
 Coefficie Standard P- Lower Upper Lower Upper
nts Error t Stat value 95% 95% 95.0% 95.0%
Intercept 381.24 502.76 0.76 0.46 -675.01 1437.50 -675.01 1437.50
DLHs per
ship 4.37 0.16 27.31 0.00 4.03 4.71 4.03 4.71

DLHs per ship Residual Plot

1500
1000
Residuals

500
0
-5002000 2500 3000 3500 4000 4500 5000 5500
-1000
DLHs per ship

Normal Probability Plot

Direct Labor-related MOH

25000
20000
15000
10000 Series1
5000
0
5 5 5 5 5 5 5 5 5 5
2. 12. 22. 32. 42. 52. 62. 72. 82. 92.
Sample Percentile

If we compare both the options above, then the direct labor hours and the direct labor cost
both have a significant impact on the direct labor related manufacturing overheads however, the
relationship and the percentage explanation of direct labor related manufacturing overheads as a
result of direct labor hours is much higher as compared to direct labor cost. The relationship
between direct labor related manufacturing overheads and direct labor hours is 99% and
percentage explanation is 98% as shown by adjusted r square value. Therefore, direct labor hours
is a better cost driver for estimating the direct labor related manufacturing overheads so we
recommend option B.
 Question 2

First of all, the regression model for Option C has been generated by taking number of
ships as independent variable and direct labor hours per ship as dependent variable. The results
of this regression model are as follows:

Regression Statistics
Multiple R 0.84
R Square 0.71
Adjusted R
Square 0.70
Standard
Error 377.74
Observations 20

ANOVA
Significan
df SS MS F ce F
6394368
Regression 1 6394368.82 .82 44.81 0.00
142690.
Residual 18 2568427.97 44
Total 19 8962796.80

Coefficie Standard P- Lower Upper Lower Upper

nts Error t Stat value 95% 95% 95.0% 95.0%
Intercept 4099.06 175.47 23.36 0.00 3730.41 4467.72 3730.41 4467.72
Ship -98.06 14.65 -6.69 0.00 -128.83 -67.28 -128.83 -67.28

Ship Residual Plot

1500
1000
Residuals

500
0
-500 0 5 10 15 20 25

-1000
Ship
 Normal Probability Plot
6000
5000
4000

DLHs per ship

3000
2000 Series1
1000
0
5 5 5 5 5 5 5 5 5 5
2. 12. 22. 32. 42. 52. 62. 72. 82. 92.
Sample Percentile

Similarly, the regression model for Option D has also been generated by taking number
of ships as independent variable and log of direct labor hours per ship as dependent variable. The
results are shown below:

Regression Statistics
Multiple R 0.96
R Square 0.93
Adjusted R
Square 0.92
Standard Error 0.06
Observations 20

ANOVA
Significanc
df SS MS F eF
225.2
Regression 1 0.71 0.71 1 0.00
Residual 18 0.06 0.00
Total 19 0.76

Coefficie Standard P- Lower Upper Lower Upper

nts Error t Stat value 95% 95% 95.0% 95.0%
238.
Intercept 4970.89 0.04 02 0.00 8.44 8.59 8.44 8.59
-
15.0
LN(Units) -0.24 0.02 1 0.00 -0.27 -0.20 -0.27 -0.20
 LN(Units) Residual Plot
0.15
0.1
0.05

Residuals
0
0.000000
-0.05 1.000000 2.000000 3.000000 4.000000
-0.1
-0.15
LN(Units)

Normal Probability Plot

8.6
8.4
8.2
LN (DLHs)

8
7.8 Series1
7.6
7.4
5 5 5 5 5 5 5 5 5 5
2. 12. 22. 32. 42. 52. 62. 72. 82. 92.
Sample Percentile

If we compare both the models, then we can see that Model D has a more significant
relationship between the log of direct labor hours and the number of ships produced. Secondly, if
we compare the intercepts and coefficients of both the models, the option D has a much higher
intercept as compared to option C. However, the coefficient, which shows the learning curve
effect, is significantly lower for option D. Therefore, Option D is a better model for predicting
the direct labor hours required for building the ships in future.
 Question 3

a).

The estimated direct labor hours per each ship and in total are as follows:

Question 2 Option D model

DLH per ship = 4970.80 - 0.25x
Ship Number DLHs per ship
21 4970.55
22 4969.55
23 4962.8
24 4898.8
25 4117.55
Total 23919.25

b).

The estimated direct labor costs for each of the five ships and in total are as follows:

Question 2 Option D model

DLH per ship = 4970.80 - 0.25x
Ship Number DLHs per ship Wages per DLH DLCs per Ship
21 4970.55 34 168998.7
22 4970.3 34 168990.2
23 4970.05 35 173951.75
24 4969.8 35 173943
25 4969.55 36 178903.8
Total 24850.25 864787.45

c).

The direct labor related manufacturing overhead costs for each of the five ships and in
total are as follows:

Question 1 Option B model

Directed Labor related MOH = 381.24 + 4.37x
Ship Number DLHs per ship MOH costs per ship
 21 4970.55 22102.54
22 4970.3 22101.45
23 4970.05 22100.36
24 4969.8 22099.27
25 4969.55 22098.17
Total 110501.79

d).

The total costs for each of the five ships and in total are as follows:

Total Costs per Ship

Ship Number DLC per ship MOH per ship Total Costs
21 168998.7 22102.54 191101.24
22 168990.2 22101.45 191091.65
23 173951.75 22100.36 196052.11
24 173943 22099.27 196042.27
25 178903.8 22098.17 201001.97
Sum 864787.45 110501.79 975289.24
Total Costs 975289.24

The total costs per ship follow the learning curve as determined in question two above.
This could be seen by the decline in the total costs per each ship as the number of ships
increases. The learning effect is reflected first in the direct labor hours and then in the MOH
costs per ship.