Objective: To advise management about whether cutting flight capacity by 11% will turn Continental into a profitable firm in the near future (within the next year). The report should include information gathered from answering the following questions.
Requirement: Read the Continental Case and answer the following questions. The case and data are on Blackboard.
1. Use the quarterly data for operating costs and the various cost drivers of costs provided by Exhibit 1 to estimate a simple regression for each cost category. Interpret the regression results and include the appropriate cost function for each cost category. Interpreting the results require you to explain the meaning behind the intercept and the coefficient for each regression. Do not estimate regression equations for fuel and depreciation.
In some cases outliers may be a factor, so you must decide what to do with these. Include in the report an exhibit that resembles the following:
# of obs used
Salaries and Wages
Important note about the use of regression: Regression (in the context of cost functions) is useful when trying to determine activities that drive costs and how costs behave. Cost drivers have already been identified for this exercise. But, you are welcome to use and identify alternative cost drivers not provided in the case. The objective of requirement #1 is for you to determine how each cost behaves relative to the given cost driver. There is no need to determine how fuel costs behave and the unit costs related to fuel because fuel costs are known. Further, because depreciation is based on accounting policy and not on activity level, this is also known, thus, there is no need to use regression to determine how these costs behave.
Important notes about outliers: Outliers for this case means “…observations that would not be useful in forecasting future expenses.” I understand that you may have learned various techniques to identify outliers in your QA course. I recommend that you not use those techniques for this case because you do not have enough data points to make the use of those techniques effective. I suggest using scatterplots to help identify outliers. If you eliminate ‘outliers’ you must have justification for doing so. A justification is not ‘we eliminated the observation because it was adversely affecting our regression model.” It is not necessary that you provide proof that your justification is a reasonable one. But, it is necessary that you provide a reasonable justification. It is highly likely that the case will not provide you with the information you will need to justify the elimination of all outliers identified. In these cases you will need to rely on resources beyond those provided in the case. Finally, do not feel the need to identify and eliminate outliers for each regression.
2. Based on your regression results, project quarterly and total costs for fiscal year 2009. Include in the report an exhibit that resembles the following:
Total Forecast 2009