Lesson 7: Forecasting and solving time series problems
Forecast a year of guest nights
SkillForecasting
Worked example
For the holiday park, (thousands, is Q1 2022). The seasonal indices are Q1 1.262, Q2 0.698, Q3 0.964 and Q4 1.076.
Forecast the guest nights for each quarter of 2025, in thousands, correct to one decimal place.
Step 1: Find t for each quarter of 2025
The data ends at Q4 2024, . So Q1 to Q4 2025 are .
Step 2: Find the trend values
:
: : :
These are deseasonalised forecasts — what an average quarter would be at each time.
Step 3: Re-seasonalise
Q1: Q2:
Q3: Q4:
Step 4: Comment on reliability
These are extrapolations of up to four quarters beyond the data. The trend is strong () and the seasonal pattern has been very stable, so they are reasonably reliable — provided nothing unusual (a cyclone, a new competitor) happens.
Answer
Forecasts for 2025: Q1 45.6, Q2 25.6, Q3 35.9 and Q4 40.7 thousand guest nights.
Try a similar one
A fresh question of the same kind, checked as you answer, with hints and a full solution.
Method
Find for each period to be forecast.
Substitute into the trend line to get the trend (deseasonalised) value.
Multiply by the season's index.
Watch out for
Using to for 2025, or .
Stopping at the trend value (36.1 for Q1) without multiplying by the index.