Time-Series Graphs and Trends
Key Terms
- Time series
- A sequence of data values measured at regular time intervals; time is always the explanatory variable on the x-axis.
- Trend
- The long-term direction of the data: upward, downward, or stationary (flat).
- Seasonal variation
- A regular, repeating pattern within a fixed time period (e.g. higher retail sales every December).
- Cyclical variation
- Longer-term rises and falls with no fixed period (e.g. economic cycles spanning several years).
- Irregular variation
- Unpredictable, one-off fluctuations caused by unusual events (e.g. natural disasters, strikes).
- Stationarity
- A time series with no long-term trend; values fluctuate around a constant level.
What is a Time Series?
A time series is a sequence of data values measured at regular time intervals (e.g. monthly, quarterly, annually). Time is always the explanatory variable and is plotted on the horizontal axis. Consecutive points are joined with line segments to reveal patterns over time.
Components of a Time Series
Seasonal Variation (S): A regular, repeating pattern within a fixed period (e.g. each year). For example, higher retail sales every December.
Cyclical Variation (C): Longer-term rises and falls not of fixed period (e.g. economic cycles of 5–10 years).
Irregular (Random) Variation (I): Unpredictable, one-off fluctuations caused by unusual events (droughts, strikes, pandemics).
Identifying Trend
To identify the trend visually: draw or imagine a smooth curve through the data, ignoring short-term ups and downs. If this smooth curve slopes upward, there is an upward trend. If it slopes downward, there is a downward trend. If it is roughly flat, the series is stationary.
Worked Example
The quarterly visitor numbers (thousands) to a Queensland national park over 2 years are shown below. Describe the components of the time series.
| Quarter | Q1 Y1 | Q2 Y1 | Q3 Y1 | Q4 Y1 | Q1 Y2 | Q2 Y2 | Q3 Y2 | Q4 Y2 |
|---|---|---|---|---|---|---|---|---|
| Visitors | 42 | 28 | 18 | 35 | 48 | 32 | 22 | 40 |
Solution: Trend: Gently upward (the values in Year 2 are slightly higher than the corresponding quarters in Year 1). Seasonal variation: Yes — Q1 (summer) has the highest visitors, Q3 (winter) has the lowest, and this pattern repeats each year. Irregular variation: Small random differences around the seasonal pattern.
Full Lesson: Time-Series Graphs and Trends
1. Why Time Series Analysis?
Many important decisions in business, government, and science depend on understanding how quantities change over time. A retailer needs to predict Christmas demand; a government health authority monitors flu case counts each week; an economist tracks quarterly GDP. Time series analysis provides tools to separate the meaningful long-term signal from seasonal patterns and random noise.
2. Constructing a Time Series Plot
Step 1: Label the horizontal axis as time (using the actual time labels: months, quarters, years). Step 2: Label the vertical axis as the measured variable with its units. Step 3: Plot each data value at its correct time point. Step 4: Join consecutive points with straight line segments. Unlike a scatterplot, always join the points in a time series plot, because the order of points matters and the line helps reveal the pattern.
3. The Four Components in Detail
Trend: Think of drawing a smooth line through the middle of the data, ignoring the ups and downs. If this line slopes upward, there is an upward (increasing) trend. If it slopes downward, there is a downward (decreasing) trend. A roughly flat line means the series is stationary (no trend). Many real datasets show a clear trend that can be modelled by a straight line.
Seasonal variation: This is a regular, predictable pattern that repeats each year (or each fixed period). For example: electricity consumption is higher in summer and winter (cooling and heating) and lower in spring and autumn. Ice cream sales peak in summer. Retail sales spike in December. Seasonal variation is predictable, which makes it useful for forecasting. The key word is regular — it repeats with approximately the same pattern each period.
Cyclical variation: Longer swings that are not of a fixed period. Economic expansions and recessions are a classic example. Unlike seasonal variation, cyclical variation is harder to predict because the duration of each cycle varies. For General Maths purposes, cyclical variation is less commonly examined than the other three components.
Irregular variation: The leftover random fluctuation after trend and seasonal effects have been accounted for. Every real dataset has some irregular variation — the data never follows a perfect mathematical formula. Irregular variation includes the effects of unusual one-off events: a factory fire, an unexpected election result, a flood. Irregular variation cannot be predicted, only acknowledged.
4. Describing a Time Series
When asked to describe a time series, address all relevant components:
- State whether there is a trend (upward/downward/stationary) and describe its approximate rate.
- State whether there is seasonal variation: does the same pattern repeat each year? Which period is consistently highest/lowest?
- Note any obvious outliers or unusual observations (irregular variation).
Use hedging language: "there appears to be", "the data suggests", "there is evidence of". Time series data is observational — we describe patterns, not prove causes.
5. Practical Example: Australian Retail Sales
Australian monthly retail trade data consistently shows: an upward trend over many years (reflecting population growth and inflation); a strong seasonal spike in December each year (Christmas shopping); a slight dip in January (post-Christmas correction). Any individual month may deviate from this pattern due to irregular events (a COVID lockdown, a natural disaster, a cost-of-living shock).
6. Key Vocabulary Summary
- Time series: data collected at regular time intervals, plotted over time
- Trend: long-term upward, downward, or stationary direction
- Seasonal variation: regular, repeating pattern within each year
- Irregular variation: random, unpredictable fluctuation
- Stationary series: no trend; values fluctuate around a constant mean
Mastery Practice
-
Fluency
The table shows quarterly revenue ($000) for a small logistics company over two years.
Quarter Q1 Y1 Q2 Y1 Q3 Y1 Q4 Y1 Q1 Y2 Q2 Y2 Q3 Y2 Q4 Y2 Revenue ($000) 85 92 78 110 91 99 84 118 Describe the overall trend in the revenue data.
-
Fluency
The table shows the number of new house building approvals (hundreds) in a coastal city, recorded each quarter over two years.
Quarter Q1 Y1 Q2 Y1 Q3 Y1 Q4 Y1 Q1 Y2 Q2 Y2 Q3 Y2 Q4 Y2 Approvals 14 22 19 12 15 23 20 13 Does this time series show seasonal variation? Justify your answer by referring to the data.
-
Fluency
The monthly electricity consumption (MWh) for a Queensland factory is shown below for one year.
Month Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec MWh 480 465 420 385 370 410 430 425 395 380 415 470 State the axes labels you would use when constructing a time series plot of this data, and identify which months show the highest and lowest consumption.
-
Fluency
In the context of a time series, define irregular variation and give one realistic example from Australian economic data.
-
Understanding
The table shows annual average petrol prices (cents per litre) in Australia over 8 years.
Year 1 2 3 4 5 6 7 8 Price (c/L) 138 145 142 155 160 158 172 179 - Describe the overall trend of the data.
- In Year 3 the price dropped slightly. Identify which component of the time series this represents.
- Would you expect this data to show seasonal variation? Explain.
-
Understanding
Monthly tourist arrivals (thousands) to a tropical island resort are recorded below for 18 months.
Month 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 Arrivals 52 48 35 30 28 38 60 72 68 42 38 55 58 54 40 34 32 44 - Identify the peak period (months) for tourist arrivals. What component does this represent?
- Compare arrivals in Month 1 and Month 13 (both January). What does any difference suggest?
-
Understanding
A student draws a time series plot and observes that the values fluctuate up and down but there is no consistent long-term upward or downward direction over 5 years.
- What term describes this type of trend?
- Despite no overall trend, the student notices values are always higher in December than in June every year. What component is this?
- In Year 3, values were much lower than expected due to a major factory closure. What component is this?
-
Understanding
The table below shows quarterly sales of winter clothing ($000) for a fashion retailer over 3 years.
Year Q1 Q2 Q3 Q4 1 42 78 95 55 2 48 85 104 62 3 53 91 112 68 - Describe the trend in Q3 sales across the three years.
- Identify which quarter consistently has the highest and which has the lowest sales. Explain in terms of time series components.
- Calculate the average annual sales for each year. Do the averages confirm the trend?
-
Problem Solving
The table shows monthly unemployment rates (%) in a regional area over 24 months (two full years).
Month 1 2 3 4 5 6 7 8 9 10 11 12 Y1 (%) 7.2 7.0 6.8 6.5 6.3 6.6 7.0 7.1 6.9 6.4 6.1 5.8 Y2 (%) 6.9 6.7 6.4 6.1 5.9 6.2 6.6 6.7 6.5 6.0 5.7 5.3 - Describe the overall trend in unemployment rates over the two years.
- Within each year, unemployment rises in Months 6–8 then falls again. What component does this represent? Suggest a possible explanation.
- Calculate the annual average unemployment rate for each year. By how many percentage points has the average fallen?
- In Month 4 of Year 2, a new major employer announced 500 new jobs in the region. Would this appear as a trend, seasonal, or irregular component? Explain.
-
Problem Solving
A supermarket chain records quarterly sales of a sunscreen product ($000) over three years. The managing director claims: "Our sales are growing steadily and there is no seasonal pattern — the product sells equally well all year round."
Year Q1 (Summer) Q2 (Autumn) Q3 (Winter) Q4 (Spring) 1 210 140 90 165 2 228 152 98 180 3 245 165 108 196 - Calculate the annual averages for each year and comment on the trend.
- Using the data, evaluate the managing director's claim about seasonal patterns. Include numerical evidence.
- Express Q3 sales as a percentage of Q1 sales in each year. What does this reveal?