Cost Estimation Question

Cost Estimation

Lecture 13

Chapter 5

Implementation Problems with Regression Data

1. Curvilinear costs

2. Outliers

3. Spurious relations

4. Assumptions

Implementation Problems, Continued. . .

1. Curvilinear costs

Relevant

Range

Problem: Attempting to fit a linear model to nonlinear data. Likely to occur near full- capacity.

Solution: Define a more limited relevant range (example: from 25 – 75% capacity) or design a nonlinear model.

Implementation Problems, Continued. . .

2. Outliers

True regression

line

Regression line

with outlier

outlier

Problem: Outlier moves the regression line.

Solution: Prepare a scatter-graph, analyze the graph and eliminate highly unusual observations before running the regression.

Implementation Problems, Continued. . .

3. Spurious relations

Problem: Using too many variables in the regression. For example, using direct labor to explain materials costs. Although the association is very high, actually both are driven by output.

Solution: Carefully analyze each variable and determine the relationship among all elements before using in the regression.

Data Problems

Missing data

Outliers

Inflation

Mismatched time periods

Practice Problems

Chapter 5

Methods to Estimate Cost Behavior

Account analysis

Engineering estimates

Statistical methods

Recap of what we covered last class

Account Analysis

Review each

account comprising

the total cost being

analyzed.

Identify each cost as

either fixed or

variable.

Activity level

Costs ($)

Fixed

Variable

Cost ($)

Activity level

Recap of what we covered last class

Example: Account Analysis

3C Cost Estimation Using Account Analysis

Costs for 360 Repair Hours

Account Total

Office Rent $3,375

Utilities 310

Administration 3,386

Supplies 2,276

Training 666

Other 613

Total $10,626

Per Repair Hour

Variable Cost Fixed Cost

$1,375 $2,000

100 210

186 3,200

2,176 100

316 350

257 356

$4,410 $6,216

12.25* Number of Repair Hours6,216 +Total Cost =

$12.25

Recap of what we covered last class

Practice Problem 1 – Account Analysis

Practice Problem 1 – Account Analysis

Practice Problem 1 – Account Analysis

Practice Problem 2 – Account Analysis

Practice Problem 2 – Account Analysis

Practice Problem 2 – Account Analysis

Statistical Cost Estimation

Statistical procedure to determine the

relationship between variables.

High-Low Method

Regression

Uses two data points.

Uses all the data points.

3C Overhead

Recap of what we covered last class

High-Low Cost Estimation

V Cost at highest activity Cost at lowest activity

Highest activity Lowest activity

F Highest activityTotal cost at highest activity level

V

or

Lowest activityTotal cost at lowest

activity level V

Recap of what we covered last class

Practice Problem 3 – High-Low Cost Estimation

Practice Problem 3 – High-Low Cost Estimation

Maintenance costs = Fixed Costs + 1(number of visitors)

Month Labor-Hours Machine-HoursOverhead Costs

1 725 1355 102687

2 715 1407 103792

3 680 1520 109915

4 740 1453 108280

5 780 1591 116229

6 755 1579 114464

7 740 1390 107022

8 725 1306 102094

9 710 1454 106439

10 795 1545 113067

11 675 1298 100755

12 710 1604 112842

Practice Problem 4 – High-Low Cost Estimation A

Practice Problem 4 – High-Low Cost Estimation

Month Labor-Hours Machine-HoursOverhead Costs

1 725 1355 102687

2 715 1407 103792

3 680 1520 109915

4 740 1453 108280

5 780 1591 116229

6 755 1579 114464

7 740 1390 107022

8 725 1306 102094

9 710 1454 106439

10 795 1545 113067

11 675 1298 100755

12 710 1604 112842

Practice Problem 4 – High-Low Cost Estimation A

Month Labor-Hours Machine-Hours Overhead Costs

1 725 1355 102687

2 715 1407 103792

3 680 1520 109915

4 740 1453 108280

5 780 1591 116229

6 755 1579 114464

7 740 1390 107022

8 725 1306 102094

9 710 1454 106439

10 795 1545 113067

11 675 1298 100755

12 710 1604 112842

Practice Problem 5 – Simple Regression

Considering the machine-hours as predictor, what is the simple regression equation?

A

Practice Problem 5 – Simple Regression

Month Labor-Hours Machine-HoursOverhead Costs

1 725 1355 102687

2 715 1407 103792

3 680 1520 109915

4 740 1453 108280

5 780 1591 116229

6 755 1579 114464

7 740 1390 107022

8 725 1306 102094

9 710 1454 106439

10 795 1545 113067

11 675 1298 100755

12 710 1604 112842

Practice Problem 5 – Simple Regression

Considering the labor-hours as predictor, what is the simple regression equation?

A

Practice Problem 5 – Simple Regression

Multiple Regression

Multiple Regression: When more than one predictor (x)

is in the model.

Equation: TC = V1 (X1) + V2 (X2) + FC

Recap of what we covered last class

Month Labor-Hours Machine-HoursOverhead Costs

1 725 1355 102687

2 715 1407 103792

3 680 1520 109915

4 740 1453 108280

5 780 1591 116229

6 755 1579 114464

7 740 1390 107022

8 725 1306 102094

9 710 1454 106439

10 795 1545 113067

11 675 1298 100755

12 710 1604 112842

Considering both labor-hours and machine-hours, what is the simple regression equation?

Practice Problem 6 – Multiple Regression A

Practice Problem 6 – Multiple Regression