Methods of cost estimation
a) Accounts classification (separating mixed costs) entails the examination of accounts and regards and classifying each item of expenditure into fixed, variable and semi variable. Although the method is quick and inexpensive and it is considerably subjective and inaccurate.
b) Industrial engineering (cost estimation and forecasting) this is considered is the most scientific method of establishing a cost standard. Work study techniques are applied to determine levels of input needed to satisfy given levels of outputs. Those, inputs are then turned into standards in order to estimate product cost in the future.
-Advantages
1. It enables an organization to determine the most effective way to apply resources.
2. Standard can be set using efficient usage.
3. There is control of operation by comparing actual results with the expected results
-Disadvantages
1. It is costly to use as it involves experts.
2. It is not effective for controlling many types of overhead costs.
3. It is not easy to apply in non-manufacturing activities since relationship between cost and output cannot be determined.
c) High – low method (used for separating cost and cost estimation and forecasting). This method examines cost at high and low levels making an assumption that the increase in cost between the two levels is directly due to the increase in activity and therefore represent the variable cost. Two previous accounting periods are chosen one with the highest activity level and other lowest activity level.
- Steps involved
i) Select highest and the lowest activity level.
ii) Select corresponding highest cost and lowest corresponding cost.
iii) Obtain the difference in cost and the difference in activity level.
iv) Divide the difference in cost by difference in activity to get the rate of variable cost.
v) Compute fixed cost by subtracting variable from total cost.
vi) Formulate linear prediction equation.
Example 3.1
Nixon an automobile technician has been operating a garage in Mombasa for the past two years. A year ago he converted part of his garage to a welding shop making and selling metal doors and windows. He had anticipated that the cost of the wielding shop would primary be final but has realized that the welding cost increased with the increase in with number of welding job assignments. The costs of welding job assignments are as follows:
Required: Formulate an equation to estimate the total cost of the wielding shop and compute the cost of undertaking 1256 assignments using:
(i) High – low method (ii) Simple linear regression method
Solution (using High-low method)
-Advantages of the high low method
1. It is easy to use.
2. The lowest and the highest item will cover the relevant range.
3. It takes into account possible extremes of cost.
-Disadvantages
1. It is not logical to use two points to represent all the points.
2. The estimated cost function poorly describes the actual cost relationship.
3. Costs are not properly matched with the independent variable.
d) Linear regression analysis is a mathematics technique from which the case function y = a + bx can be derived. It is commonly known as the method of mixed squares because the mathematics that is used tries to equate the sum of squared distances above the line of best fit to those below it. A regression equation utilizes historical data to estimate the relationship between the dependent variable y and one or more independent variables. The method tries to estimate mathematically the line of best fit.
In obtaining the values of a and b the following equations are used:
Example 3.2
The following data relates too Warp limited which process a single data type of a chemical.
Overhead processing cost for the last 13 accounting periods each of four weeks has been as follows.
- Coefficient of correlation (r) Correlation means a relationship between two variables. The strength of linear relationship between two variables is determined by the use a coefficient of correlation. Two pairs of data may be positively or negatively correlated as shown in Figures 3.1 and 3.2.
Correlation coefficient range between 1 and -1. The closer the correlation coefficient to positive one the strong is the positive correlation and the more close correction coefficient is to negative 1 the strong is the negative correction. Correlation coefficient is given by the following formula.
- Coefficient of determination (r
2) this is used to determine how much of the change in y is a s a result of change in x. It is obtained by squaring coefficient of correction. From the coefficient calculated, above the coefficient of determination is (0.9352)
2 = 0.87 or 87%. It means that 87% of the change in overhead cost is due to output other factors cause 13% change in the overhead cost.
e) Scatter graph is a graphical method of data estimation. Two pairs of data are plotted us single points on a graph and then joined using the line of best fit. The gradient of the line of best fit is the rate of variable cost and the y intercept represent fixed cost as indicated in Figure 3.3. Compared to other methods it is inaccurate and subjective.
f) Learning curve theory also referred to as improvement curve theory. It occurs when new methods are introduced or new products are made. It is based on proposition that as workers continue doing the same job over and over again experience is gained the time taken to complete a job or a unit of a product reduces by a certain constant referred to as learning coefficient. It is therefore possible to determine time taken or time to be taken to complete certain number of units and hence the cost of labour learning effect and effect labour cost.
marto answered the question on
February 15, 2019 at 08:33