Time Series

Time series data is everywhere in finance and economics. Daily share prices, monthly inflation figures, quarterly GDP, annual company profits and hourly exchange rates are all time series.

Analysing time series helps identify patterns and make forecasts. Common features include trend, the long-term direction of the data; seasonality, regular patterns that repeat at fixed intervals, such as higher retail sales before holidays; cycles, longer fluctuations linked to economic conditions; and random noise, irregular movements that cannot be explained.

For example, an analyst studying a retailer's monthly sales over ten years might identify steady growth, a sharp rise every December and a dip during recessions.

Special statistical methods are used for time series because observations are usually related to previous ones. Today's share price, for instance, is closely linked to yesterday's. Methods such as moving averages, exponential smoothing and more advanced models are used to describe patterns and make forecasts.

Time series analysis has limitations. Past patterns may not continue, especially when structural changes, such as new technology or policy shifts, alter the underlying behaviour.