Time series graphs
- A time series is data recorded at regular intervals over time, such as monthly sales. Plot it with time on the horizontal axis and join the points with straight lines.
- The trend is the general direction over time. Seasonal variation is a pattern that repeats, such as higher ice cream sales every summer. Random variation is unpredictable.
Moving averages
- Moving averages smooth out seasonal variation to show the trend. Use the same number of points as the length of the pattern: 4 points for quarterly data and 12 for monthly data.
- Example: quarterly sales 10, 14, 12, 16, 18. The first 4-point moving average is (10 + 14 + 12 + 16) ÷ 4 = 13. The next is (14 + 12 + 16 + 18) ÷ 4 = 15.
- Plot each moving average at the middle of the time period it covers, then draw a trend line through them.
Seasonal variation and predictions
- Seasonal variation for a point = actual value − trend value.
- The mean seasonal variation for a season (such as quarter 1) is the average of all its seasonal variations.
- To predict: read the trend line forward, then add the mean seasonal variation. If the predicted trend for next quarter 3 is 40 and its mean seasonal variation is −5, the prediction is 35.
Key terms
- Time series
- Data recorded at regular intervals over time.
- Trend
- The general direction of data over time.
- Seasonal variation
- A pattern that repeats over a regular period.
- Moving average
- An average of a set number of consecutive values, moved along one at a time.
- Trend line
- A line drawn through the moving averages to show the trend.
- Mean seasonal variation
- The average difference between actual values and the trend for one season.