Week 2

Question 1

Cost estimation

Answer 1

Development of a relationship between a cost object and its cost drivers to
predict the cost. It facilitates strategic management in three ways:

1. Predict future costs using previously identified activity-based, volume-based, structural or executional cost drivers;
2. Identify the key cost drivers for a cost object;
3. Cost drivers and cost-estimating relationships are useful in planning and decision-making.

Question 2

The high-low method

Answer 2

Easy to apply, but less accurate than regression analysis, and uses algebra
to determine a unique estimation line between representative high and low points in the data.

Question 3

Regression analysis

Answer 3

Statistical method for obtaining the unique cost-estimating equation that
best fits a set of data. To fit the data, the sum of the squares of the estimation errors are minimized.

Each error is the distance measured from the regression line to one of the data points. Since it systematically minimizes the estimation error, it is also called the least squares regression method.

Question 4

There are two types of variables

Answer 4

1. The dependent variable, which is usually the cost to be estimated, or also revenue or another type of financial or operating data like labour hours, cash flow;
2. The independent variable, which is the cost driver, used to estimate the value of the dependent variable. If one independent variable is used, it is a simple regression. Otherwise, if two or more variables are used, it is a multiple regression.

Question 5

Evaluating a Regression Analysis

Answer 5

There are four primary values associated with evaluating a regression analysis:
1) R-Squared
2) T-value
3) Standard error of estimate
4)

Question 6

Evaluating a Regression Analysis

Answer 6

There are four primary values associated with evaluating a regression analysis:

1) R-Squared
2) T-value
3) Standard error of estimate
4) P-value

Question 7

Time-series regression

Answer 7

The application of regression analysis to predict future amounts, using prior period’s data.

Question 8

Cross-sectional regression

Answer 8

Estimates cost for a cost object based on information on other cost objects and variables, where the information for all variables is taken from the same
time.

Question 9

Implementation problems with linear regression

Answer 9

Linear regression estimates are unreliable when the data relationships are nonlinear. It most often happens because of time-series patterns (e.g. seasonality), outliers, or data shifts. While the high-low method cannot be adapted, regression analysis can. Depending on the situation, different actions
can be undertaken:

1) Seasonality: if present, the most common methods to deal with it are:
- Use of a price change index to adjust the values of each variable to some common time;
- Use of a trend variable, which takes on the values 1,2,3… for each period;
- Replacement of the original values of each of the variables with the first differences;
2) Outliers, which may be resolved using a dummy variable;
3) Data shift, which can be adjusted with a dummy variable to indicate the periods before and after the shift.

Question 10

Linear curve analysis

Answer 10

Nonlinear cost behaviour is demonstrated by the application of the learning curve analysis, a systematic estimation of costs when learning and improvement in production over time is present.

The associated learning rate is the percentage by which average or total production time decreases as output doubles. The learning rate is usually based on an assumed doubling of output.

Question 11

Limitations of learning curves

Answer 11

• More relevant for labour-intensive contexts that involve repetitive tasks;
• Learning rate is assumed to be constant;
• The estimated learning curve could be unreliable, as the observed change in productivity in the data could be associated with factors besides learning.