Regression analysis

In this lesson

1. 0:00 Introduction to regression analysis
2. 0:50 Regression analysis vs the Hi-Lo method
3. 2:02 Understanding the straight line equation
4. 3:25 Applying linear equations to cost behaviour
5. 5:09 Identifying cost types from graphs
6. 7:00 Deriving the regression equation
7. 11:02 Calculating the regression coefficients
8. 18:30 Using the equation for cost prediction
9. 19:05 Understanding correlation coefficient
10. 24:02 Calculating the correlation coefficient
11. 25:40 Coefficient of determination explained
12. 27:38 Limitations of regression analysis
13. 29:08 Practice question on correlation coefficient

About this lesson

Unlike Hi-Lo, which uses only the highest and lowest data points, regression analysis incorporates all your data to establish the relationship between two variables—typically output and total costs. You'll discover how to derive the equation of a straight line (y = a + bx) where the intercept represents fixed costs and the gradient represents variable cost per unit, allowing you to predict future costs with greater accuracy. The lesson walks you through the practical steps of regression analysis, starting with scatter graphs and lines of best fit, then moving to the formulas for calculating the constants a and b. You'll work through a complete worked example using real data, learning how to set up your calculations systematically and interpret the results. The session explains how to identify different cost types—fixed, variable, and semi-variable—within the context of linear equations, helping you apply this technique to real management accounting scenarios. Beyond calculating the equation, you'll learn to assess how reliable your predictions are by calculating the correlation coefficient (r) and the coefficient of determination (r squared). These measures tell you the strength of the relationship between your variables and what proportion of cost changes are actually driven by output changes. The lesson also covers the key limitations of regression analysis, including its assumption of linear relationships and its reliance on historical data, ensuring you understand when and how to use this forecasting technique appropriately in your Performance Management studies.

Duration: 30m