Time series analysis examines patterns in data over time. It relies on identifying trends, measuring past patterns to forecast the future, and decomposing time series into four main components: secular trends, cyclical movements, seasonal variations, and irregular variations. Secular trends represent long-term direction, while cyclical and seasonal variations have recurring patterns over different time scales. Various techniques can depict trends and identify variations, including freehand drawing, semi-averages, moving averages, least squares, and exponential smoothing.