Regression Analysis Method of Cost Estimation
It includes estimating the cost function by utilizing past data or the dependent and the independent variables. Hence the cost function is based upon the regression of the relevant/related variables. The cost function will depend upon the relationship among the independent variable and the dependent variable.
- The dependent variable will constitute the relevant cost that may be service, variable cost, overhead cost. And
- The independent variable will be the cost drivers whereas the cost drivers will be units of labour, labour hours or raw materials, units of output.
In analysis of regression, a regression model of the form y= a + bx for an easy regression is obtained. Used for a multiple regression, then a regression model of the form Y = a + b_{1}x_{1} +b_{2}x_{2} + b_{n}x_{n} is acquired where a is fixed cost, x_{1},x_{2},x_{n} are cost drivers x_{1},x_{2},x_{3} upto x_{n}.
b_{1},b_{2}, b_{n} are changes in cost along with the change in value of cost driver that is variable cost per unit of change in x_{1},x_{2},x_{n }y is the dependant variable (Total cost)
Note here a simple regression produces a cost function of the form as y = a + bx so now we only contain only one variable cost per unit (b) and only one independent variable cost driver x..
Conversely, a multiple regression creates a cost function of the form as y = a + b_{1},x_{1}+ b_{2}, x_{2} + b_{n},x_{n} so now we have some variable costs per unit (b_{1},b_{2},b_{n}) and some independent variables (x_{1},x_{2},x_{n}).