Seasonal effects and forecasting

← Previous3 of 4Next →

What this lesson covers

  • Seasonal effect = data − trend.
  • A forecast extends the trend, then adds the seasonal effect.
  • Forecasts get less reliable the further ahead they go.
Worked example

Seasonal effects for Kōwhai Café

  • Data minus centred trend.
QuarterDataTrendEffect
Q35056−6
Q47058+12
Q14860−12
Q26862+6
Worked example

Forecasting Y3 Q1

  • The trend rises 2 a quarter, and Y3 Q1 is three quarters after Y2 Q2.
  1. trend = 62 + 3 × 2 = 68
  2. seasonal effect = −12
  3. forecast = 56 (that is, $56 000)

Common mistakes

  • Forgetting the seasonal effect.
    Add it to the trend every time.
  • Forecasting years ahead with confidence.
    Say that the trend may change.

Your turn

  • 1. Forecast Y3 Q2 for Kōwhai Café.
Show answer

1. trend 70, plus 6, gives 76, or $76 000

Summary

What to take away

  • Trend plus seasonal effect gives the forecast.
  • Keep forecasts short range.

More in AS 91580

Features of a time seriesNotesMoving averages and the trendNotesSeasonal effects and forecastingNotesReporting on a time seriesNotes

Is your child on track?

Take the free placement test