{[ promptMessage ]}

Bookmark it

{[ promptMessage ]}

# 16_Overheads - Objectives for today What are some questions...

This preview shows pages 1–5. Sign up to view the full content.

ACC 333 (Farrell), Fall 2011, Class 16 1 ACC 333 (Farrel ) Fal 2011, Class 16 1 Objectives for today What are some questions regression can help us with? Cost estimation using regression analysis – Simple regression and multiple regression • General form of regression equations • Estimated equations from a regression analysis • Goodness of fit • Hypothesis testing – Incorporating and interpreting dummy variables ACC 333 (Farrel ) Fal 2011, Class 16 2 What are some questions regression can help us with? Regression can help with almost any decision for which we have observations from a population or a subsample For example, if we're analyzing past performance or forecasting future performance , we can explain or predict : – sales – input costs – overhead costs – nonfinancial measures of performance

This preview has intentionally blurred sections. Sign up to view the full version.

View Full Document
ACC 333 (Farrell), Fall 2011, Class 16 2 ACC 333 (Farrel ) Fal 2011, Class 16 3 What are some questions regression can help us with? (continued) If we're focusing on costs , regression can help us answer questions like these: How will spending vary across different alternatives? What are the drivers of particular costs? Can we identify the variable and fixed components of costs for a particular cost objective? ACC 333 (Farrel ) Fal 2011, Class 16 4 Example: Focus on costs 0 500 1000 1500 2000 2500 0 50 100 150 200 250 Number of Suppliers Costs (in thousands) To solve, we can use: High-Low Method (i.e., two equations and two unknowns) (circled points) Regression Analysis (all points or a subset, if sampling) Total costs = Total fixed costs + (Variable cost/unit x Units of cost driver)
ACC 333 (Farrell), Fall 2011, Class 16 3 ACC 333 (Farrel ) Fal 2011, Class 16 5 Cost estimation using regression analysis TC = FC + [VC x X] where: TC = Total costs FC = Total fixed costs (more precisely, costs that don't vary with changes in the cost driver) VC = Variable cost / unit X = Units of cost driver (e.g., number of suppliers) slope intercept This is the equation for a line ACC 333 (Farrel ) Fal 2011, Class 16 6 Cost estimation using regression analysis (continued) Repeating our formula: TC = FC + [ VC x X ] TC represents expected (estimated) total costs, which we can compute for a given level of X by plugging in the values from our regression formula If we compare expected total costs to actual total costs for that level of X, we will be off slightly, so the formula becomes: Expected TC = FC + [ VC x X ] + random variation

This preview has intentionally blurred sections. Sign up to view the full version.

View Full Document
ACC 333 (Farrell), Fall 2011, Class 16 4 ACC 333 (Farrel ) Fal 2011, Class 16 7 Cost estimation using regression analysis (continued) And this becomes the formal form of a simple regression equation that you see in statistics: Y = ß 0 + ( ß 1 * X ) + where: Y = Dependent variable (DV) (e.g., Total costs) ß 0 = Intercept (e.g., Total fixed costs); sometimes denoted ß 1 = Slope coefficient (e.g., Variable cost / unit) X = Independent/explanatory variable (IV) (e.g., units of cost driver) = Random error ACC 333 (Farrel ) Fal 2011, Class 16 8 Simple linear regression Simple regressions include only one IV Using sample output (Neidless Markop, #1), let's examine: Estimated regression equation Goodness of fit:
This is the end of the preview. Sign up to access the rest of the document.

{[ snackBarMessage ]}

### Page1 / 16

16_Overheads - Objectives for today What are some questions...

This preview shows document pages 1 - 5. Sign up to view the full document.

View Full Document
Ask a homework question - tutors are online