Lecture 2: Cost estimation

Question 1

Name two methods for estimating variable and fixed costs?

Answer 1

1. High-low method
2. Regression analysis

Question 2

Name 5 steps in estimating cost function using quantitative analysis?

Answer 2

1. Choose cost object to be predicted (dependent variable)
2. Identify cost driver (independent variable)
3. Collect data on both
4. Estimate cost function using high-low method or regression analysis
5. Evaluate cost driver of estimated cost function

Question 3

What is the high-low method?

Answer 3

uses highest and lowest observed values of the cost drivers and then:

Slope: (cost highest observation – cost lowest observation) / highest observation – lowest observation

Intercept = Cost highest observation – slope * highest observation cost driver

Question 4

Name two advantages of the high-low method?

Answer 4

• Simple to calculate
• Provides first overview of relationship between cost and cost driver

Question 5

Name two limitations of the high low method

Answer 5

1. Inefficient use of cost information
2. Highly sensitive to outliers at highest and lowest observation of the cost driver

Question 6

What is linear regression?

Answer 6

statistical method measuring average change in dependent variable associated with a unit change in one or more independent variables
- Simple linear regression uses only one (at least two) independent variables
- Function is estimated by minimizing OLS

Question 7

Name 5 ways in which a regression analysis can be evaluated?

Answer 7

1. Goodness of fit (R^2)
2. Standard error
3. T-value
4. P-value
5. Economic probability

Question 8

What is goodness of fit?

Answer 8

indicates explanatory power of regression model (to what extent can change in DV be explained by change in IV

Question 9

What is the standard error?

Answer 9

measures dispersion of actual observations around regression line and thus gives measure for accuracy of regression estimate

Question 10

What is the t-value?

Answer 10

measures validity of IV in predicting DV. Divide coefficient by standard error. Higher t-value is better

Question 11

What is the p-value?

Answer 11

a translation of t-value into probability that observed relationship is due to change

Question 12

What is economic plausibility?

Answer 12

Requires subjective assessment of cause-effect relationship

Question 13

Name 2 advantages of regression analysis?

Answer 13

• Provides objective measure
• Uses information from all available observations

BUT: CORRELATION DOES NOT EQUAL CAUSATION