Skip to content
Marlbridge

Practice Questions

Time Series and Moving Averages: O Level Statistics 4040 and IGCSE 0479 Practice Questions

Practice questions with worked answers on moving averages, trend lines and seasonal components: O Level Statistics 4040 Topic 10 and IGCSE 0479 Topic 12.

Subject
Statistics
Level
O LEVELS, IGCSE
Topic
Topic 10 – Time Series
Updated

Aligned to Cambridge O Level IGCSE Statistics (4040, 0479), 2025-2027. Official specification (O Level) ; Official specification (IGCSE) .

Syllabus page (what it covers and how it is assessed): Cambridge O Level Statistics; Cambridge IGCSE Statistics.

Syllabus topics and points this page is mapped to

4040

  • 10.1 Understanding of trend
  • 10.2 Understanding of seasonal variation

0479

  • 12 Time series (topic-level mapping)

"Topic-level mapping" means this page belongs to that topic; it does not claim to cover every point in it.

Found an error? Report a correction.

Want a teacher for this? Marlbridge teaches Statistics online. Request a free trial class.

These are original questions written for Marlbridge, for revision and practice on this content. They are not reproduced past-paper questions, and they do not replicate the exam’s exact structure, question count or mark tariffs. Use them alongside the official past papers from Cambridge.

A time series question asks you to smooth data with moving averages, find seasonal components (actual value minus trend) and use a trend line to predict a future value. In Cambridge O Level Statistics 4040 this is Topic 10 and in IGCSE Statistics 0479 it is Topic 12; the methods below follow both syllabuses.

Study the method first: time series study guide. Related: time series revision notes.


Questions

1. A school café records its sales (in hundreds of rupees) each term for three years.

Autumn Spring Summer
Year 1 34 25 19
Year 2 38 27 22
Year 3 41 31 25

(a) Explain why a 3-point moving average is appropriate for these data. [1]

(b) Calculate all the 3-point moving averages. [3]

(c) Explain why these moving averages do not need to be centred. [1]

(d) Calculate the seasonal component for each term. [3]

(e) A trend line is drawn through the first and last moving averages. Use it, with your answer to (d), to predict the café’s sales in the Autumn term of Year 4. [3]

(f) State one assumption made in your prediction. [1]

2. A household’s quarterly electricity bills (in hundreds of rupees) are:

Q1 Q2 Q3 Q4
Year 1 120 95 80 130
Year 2 128 99 86 139

(a) Calculate the five 4-point moving averages. [2]

(b) Calculate the four centred moving averages and state the quarter each belongs to. [3]

(c) For each of these four quarters, calculate the difference between the bill and the centred moving average. [2]

(d) Explain why two years of data are not enough to give a reliable seasonal component for each quarter. [1]

3. A shop’s quarterly sales (in thousands of rupees) are plotted on a time series graph, and a trend line is drawn by eye through the moving averages. The trend values read from the line are shown with the actual sales.

Year 1 Q1 Q2 Q3 Q4 Year 2 Q1 Q2 Q3 Q4
Sales 58 49 42 60 63 54 50 65
Trend value 50.0 51.5 53.0 54.5 56.0 57.5 59.0 60.5

(a) Calculate the seasonal component for each quarter. [3]

(b) The trend line rises by 1.5 each quarter. Predict the shop’s sales in Year 3 Q3. [2]

(c) Explain why a prediction for Year 5 Q3 made in the same way would be less reliable. [1]

4. For each data set, state the number of points you would use for a moving average and whether the moving averages would need to be centred.

(a) The number of visitors to a museum, recorded on each of its 5 opening days a week for six weeks. [1]

(b) A company’s monthly sales over four years, which follow a yearly pattern. [1]

(c) State one reason for finding moving averages of a time series. [1]


Answers

1. (a) There are three terms in each yearly cycle, so each moving average covers one full cycle and the seasonal pattern is averaged out [1].

(b) (34 + 25 + 19) ÷ 3 = 26.0; (25 + 19 + 38) ÷ 3 = 27.3; (19 + 38 + 27) ÷ 3 = 28.0; (38 + 27 + 22) ÷ 3 = 29.0; (27 + 22 + 41) ÷ 3 = 30.0; (22 + 41 + 31) ÷ 3 = 31.3; (41 + 31 + 25) ÷ 3 = 32.3 [3] (2 marks for five or six correct, 1 mark for three or four).

(c) With an odd number of points, the middle of each group is an actual term (for example, the first average belongs to Year 1 Spring), so it can be plotted against that term directly [1].

(d) Differences (sales − moving average):

  • Autumn: Year 2 38 − 28.0 = 10.0; Year 3 41 − 31.33 = 9.67; mean +9.83
  • Spring: Year 1 25 − 26.0 = −1.0; Year 2 27 − 29.0 = −2.0; Year 3 31 − 32.33 = −1.33; mean −1.44
  • Summer: Year 1 19 − 27.33 = −8.33; Year 2 22 − 30.0 = −8.0; mean −8.17

[1] for the method (sales − moving average), [1] for the differences, [1] for the three means.

(e) The trend line passes through (Year 1 Spring, 26.0) and (Year 3 Spring, 32.33): a rise of 6.33 over 6 terms, about 1.06 per term [1]. Year 4 Autumn is 2 terms after Year 3 Spring, so the trend value is about 32.33 + 2 × 1.06 = 34.4 [1]. Prediction = 34.4 + 9.83 ≈ 44.3, about Rs 4 430 [1].

(f) Any one: the trend continues in a straight line beyond the data; the seasonal pattern stays the same in Year 4; there are no unusual events (such as a change in prices or school closure) [1].

2. (a) (120 + 95 + 80 + 130) ÷ 4 = 106.25; then 108.25, 109.25, 110.75, 113.0 [2] (1 mark for three or four correct).

(b) (106.25 + 108.25) ÷ 2 = 107.25 (Year 1 Q3); (108.25 + 109.25) ÷ 2 = 108.75 (Year 1 Q4); (109.25 + 110.75) ÷ 2 = 110.0 (Year 2 Q1); (110.75 + 113.0) ÷ 2 = 111.875 (Year 2 Q2) [3] (1 mark for the values, 1 mark for the method, 1 mark for the quarters).

(c) Year 1 Q3: 80 − 107.25 = −27.25; Year 1 Q4: 130 − 108.75 = +21.25; Year 2 Q1: 128 − 110.0 = +18.0; Year 2 Q2: 99 − 111.875 = −12.875 [2] (1 mark for two correct).

(d) Each quarter has only one difference, so its “average” rests on a single year and may be distorted by an unusual year [1].

3. (a) Differences (sales − trend value):

  • Q1: 58 − 50.0 = 8.0; 63 − 56.0 = 7.0; mean +7.5
  • Q2: 49 − 51.5 = −2.5; 54 − 57.5 = −3.5; mean −3.0
  • Q3: 42 − 53.0 = −11.0; 50 − 59.0 = −9.0; mean −10.0
  • Q4: 60 − 54.5 = 5.5; 65 − 60.5 = 4.5; mean +5.0

[1] for the method, [1] for the eight differences, [1] for the four means.

(b) Year 3 Q3 is 3 quarters after Year 2 Q4, so the trend value is 60.5 + 3 × 1.5 = 65.0 [1]. Prediction = 65.0 − 10.0 = 55.0, about Rs 55 000 [1].

(c) It extrapolates much further beyond the data, so it relies for longer on the trend staying linear and the seasonal pattern staying the same [1].

4. (a) 5-point moving averages (one full week); 5 is odd, so no centring is needed [1].

(b) 12-point moving averages (one full year); 12 is even, so the averages fall between months and must be centred [1].

(c) Any one: to smooth out the seasonal variation and random fluctuations; to show the trend; to give values from which seasonal components can be calculated [1].


Where marks are usually lost

  • In 1(b), rounding early. Keep 27.33 and 31.33 in the working even if you quote 27.3 and 31.3.
  • In 1(d), subtracting the wrong way round, which flips every sign.
  • In 1(e), adding the seasonal component to the last actual Autumn value (41) instead of to the trend value.
  • In 2(b), attaching the centred averages to the wrong quarters. The first centred value belongs to the third quarter of the data.
  • In 3(b), counting quarters from the last actual value rather than from the last trend value, or forgetting to subtract a negative seasonal component.

Frequently asked questions

How do you choose how many points to use in a moving average? Use the number of values in one full cycle of the data: 3 for three terms a year, 4 for quarterly data, 12 for monthly data with a yearly pattern. Each average then covers exactly one cycle, so the seasonal pattern is averaged out.

When do moving averages need to be centred? When the number of points is even. A 4-point average falls between the second and third values, so you average consecutive pairs of 4-point averages to place each centred value on an actual quarter.

How is a seasonal component calculated? For each value, subtract the trend value (moving average or trend-line reading) from the actual value, then take the mean of these differences for each season. A positive component means that season is usually above the trend.

How do you predict a future value from a time series? Read or extend the trend line to the future time point, then add that season’s mean seasonal component. The prediction assumes the trend stays linear and the seasonal pattern does not change.

Next steps

Get free revision emails (optional)

Occasional emails with practice questions, worked explanations and links to free resources for the qualification and subjects you choose. No spam, and you can unsubscribe from any email. The free tools on this site never need an email.

Subjects (optional, up to 6)

Choose a qualification to see its subjects.

Related resources

Related articles

Studying this with a teacher

Working through Statistics O LEVELS, IGCSE?

This page is free and stays free. If you would rather be taught it, Marlbridge runs Statistics classes one-to-one and in small groups of up to 15, online in your own time zone. The first trial class is free. WhatsApp replies within an hour (8am–11pm Pakistan time, every day); email the same day.