Sales forecasting and time-series analysis: moving averages, identifying the trend, seasonal and cyclical variation, and the uses and limitations of extrapolation.

What this dot point is asking

WJEC Unit 3 expects you to forecast sales from past data using time-series analysis: calculating moving averages, identifying the trend, recognising seasonal and cyclical variation, and judging the uses and limits of extrapolation. This is quantitative and practical, so the marks reward correct calculation, identifying the trend, and a realistic view of how reliable a forecast is.

The answer

Why firms forecast sales

Time-series analysis and the trend

A time series is data recorded over successive periods (monthly or quarterly sales). Time-series analysis separates the data into components:

• Trend - the underlying long-term direction (rising, flat or falling).
• Seasonal variation - regular, predictable swings within a year (higher sales at Christmas, lower in spring).
• Cyclical variation - longer swings linked to the economic cycle (boom and recession).
• Random variation - irregular, one-off effects (a strike, unusual weather).

Identifying the trend is the goal, because seasonal and random swings can hide the real direction of sales.

Moving averages

Extrapolation: uses and limits

Extrapolation projects the identified trend forward to forecast future periods. It is quick, cheap and useful for planning in a stable market. But it rests on a strong assumption - that the past trend continues - so it is unreliable when the market changes (new competitors, a recession, a fashion shift) and ignores one-off events. The further ahead a forecast reaches, the less reliable it becomes. Firms therefore treat extrapolation as a starting point, combining it with market research and judgement.

Examples in context

Example 1. A seasonal retailer. A garden centre's sales peak in spring and summer and fall in winter every year. The raw monthly figures swing wildly, but a 12-month moving average smooths the seasonal pattern to reveal whether the underlying trend is up or down. This lets the owner plan staffing and stock for the predictable seasonal peaks while tracking the real long-term direction. The example shows time-series analysis separating season from trend.

Example 2. Extrapolation breaking down. A firm extrapolated a steady upward sales trend to plan a big expansion, but a new competitor and an economic downturn broke the trend and sales fell. The forecast failed precisely because extrapolation assumes the past continues. The example illustrates the central limitation: extrapolation is only as good as the assumption that conditions stay the same, which is why firms combine it with market research and judgement.

Try this

Q1. Define the term moving average. [2 marks]

• Cue. A succession of averages of a fixed number of consecutive periods, recalculated as the window moves, used to smooth out short-term fluctuations and reveal the underlying trend.

Q2. Sales over four quarters were 30, 50, 70 and 90 (£000). Calculate the four-quarter moving average. [2 marks]

• Cue. Total = 30 + 50 + 70 + 90 = 240. Moving average = 240 / 4 = £60,000.