Home Business & Finance Executive finance and strategy

# (c) Extrapolation methods

##### Informal extrapolation

This is what we often do with budgets. As an example, if sales are growing at 11 per cent, variable costs will be up 11 per cent - at this level of growth, fixed costs should be just that - fixed. This approach may work well. As with any forecast, the rationale should be explained to everyone using the forecast and also the assumptions tested.

FIGURE 10.1 Week-by-week net sales of a restaurant business

##### Graphical extrapolation

Maybe too obvious for accountants - particularly management accountants! But graphs are a blatant way of getting messages across. Yes, the scales can be manipulated, but the 'honest truth' of past performance and trends can be clearly revealed.

##### Graphs of selected data

The example shown in Figure 10.1 (given as an example of graphical reports in Chapter 9) shows week-by-week net sales of a restaurant business. Series 1 (first year) shows slow growth, series 2 (second year) decline, series 3 (third year) further decline. This business went bust, which, with that wonderful attribute, hindsight, is probably very obvious, as is the fact that it was being forecast.

The investors, optimistic folk, kept grasping at signs of an upturn - and the problem of graphs - over which period do you prepare the graph? Figure 10.2 shows a selection of the same data for the first months of each

FIGURE 10.2 A selection of the same data for the first months of each year

year - and lo and behold, while sales were falling, overall the trends are all upwards! This is just one of the issues of relying on graphs.

However, graphs are simple to compile and can be very powerful indicators of past success (or failure) and good indicators of trends.

##### Extrapolate using arithmetic

There is the simple forecasting spreadsheet function, as demonstrated in Figure 10.3. This is pure extrapolation based on past data - so you have to question how predictive the arithmetic really is. It will depend on the accuracy of the past data and how representative this might be of the future.

FIGURE 10.3 XLS forecast function

The formula in G4 is = FORECAST(G2,B3:F3,B2:F2). This gives the value to be expected in G3 for month 6.

The forecast for month 6 could be arrived at using a trend line (Figure 10.4). As this example has very few data, the trend figure for month 6 is higher -45 units.

In summary, extrapolating (to infer or estimate by extending or projecting known information) is often a very useful exercise, but it is just that - inferring or estimating.

Found a mistake? Please highlight the word and press Shift + Enter

Subjects