Time Series · Part 2 of 2
6 exam-style questions with model answers, plus 8 quick multi-choice questions — every question on this part of the standard, grouped by the 2 pages of notes they come from.
Write a full answer before you reveal the model one. That comparison is where the learning happens.
Three quarterly seasonal effects from an additive model are +5 200, −2 800 and −6 100. Find the fourth, and explain what a seasonal effect of +5 200 means.
A student's report says 'the Q4 seasonal effect is 12 400.' Explain what is missing from this statement and rewrite it properly for a series measuring quarterly retail sales in dollars.
A student deseasonalises a quarterly series of employment figures and finds the deseasonalised plot still shows a clear repeating pattern, with every Q1 sitting above the neighbouring points. Discuss what this indicates and how she should respond.
A trend is projected to be 8 400 units for the next quarter, whose seasonal effect is −1 250. Calculate the forecast and state one assumption it relies on.
A student forecasts quarterly café sales three years ahead from a five-year series. Explain why this forecast should not be relied on.
A student forecasting quarterly building consents in a region produces a figure and writes: 'The forecast for Q1 2025 is 842 consents. This assumes the trend continues and the seasonal pattern stays the same.' Evaluate this and explain how it could be improved to Excellence standard.