PPT_ch03_Management_Accounting_5e.ppt

Cost estimation (cont.) Regression analysis A statistical technique used to estimate the relationship between a dependent variable (cost) and independent variables (cost driver) The line of best fit makes deviations between the cost line and the data points as small as possible Simple regression involves estimating the relationship between the dependent variable (Y) and one independent variable (X) Y = a + bX More accurate than high–low method as it makes use of all data and has statistical properties that allows inferences to be drawn between cost and activity levels 3-* Copyright ? 2009 McGraw-Hill Australia Pty Ltd PowerPoint Slides t/a Management Accounting 5e by Langfield-Smith Prepared by Kim Langfield-Smith (cont.) Cost estimation 3-* Copyright ? 2009 McGraw-Hill Australia Pty Ltd PowerPoint Slides t/a Management Accounting 5e by Langfield-Smith Prepared by Kim Langfield-Smith Regression analysis Multiple regression estimates a linear relationship between one dependent variable and two or more independent variables Y = a + b1X1 + b2X2 The regression line can be evaluated using several criteria: Economic plausibility—does the regression line make sense? Goodness of fit—how well does the line fit the data points? Copyright ? 2009 McGraw-Hill Australia Pty Ltd PowerPoint Slides t/a Management Accounting 5e by Langfield-Smith Prepared by Kim Langfield-Smith 3-* Practical issues in cost estimation Data collection problems Missing data Outliers—extreme observations of activity or costs Mismatched time periods for dependent and independent variables Trade-offs in choosing the time period—the number of observations compared to the reliability of past data points as predictors of future cost behaviour Allocated fixed costs may be misleading Inflation may cause past cost data to be less relevant 3-* Copyright ? 2009 McGraw-Hill Australia Pty Ltd PowerPoint Slides t/a Management Accounting 5e by Langfield-Smith Prepared by Kim Langfield-Smith (cont.) Pra