Time series data is a chronological sequence of a series of values of a variable. In other words, the data is time-stamped. It can be a discrete or continuous time series. Discrete-time series are observations taken at specific times while continuous time series are observations that are made continuously over time.
It can be univariate if data on only one variable is observed or multivariate series with multiple variables are observed. Based on the time interval for which the data is observed they can be called daily/weekly/monthly/quarterly/half-yearly, annually, quinquennially, and decadal and centennial.
For example data on daily/weekly/monthly return of stocks, quarterly/half-yearly/annual data on GDP. NSSO Data on monthly expenditure and census etc.
Components of Time Series
Time series can be decomposed into four main components; Trend, Cyclical, Seasonal and Irregular components.
Trend is the tendency of the time series to increase, decrease or remain stable over a period of time, trend is the long-term movement of the time series. For example, we may observe increasing values and an upward trend in the demand of umbrellas during summer and rainy months and lower demand and thus, downward trend in the months not associated with heat and rain.
Seasonality is the predictable behavior of the time series. This interval could be annual, monthly, weekly, daily, etc. For example, sweaters are in demand as winter approaches and it weakens as the summer approaches. Like the harvest season, the presence of seasonality is presumed beforehand in time series. That’s why the data is detrended and deseasonalized before running for regression.
A cyclic pattern exists in the series due to business cycles: expansion, peak, recession, depression, trough, recovery, and depression. Each of the business cycles may last more than 2 years. This happens due to fluctuations in aggregate demand and giving rise to disequilibrium in the economy.
Irregular components are attributed to highly random events with no particular patterns. These variations are caused by random incidents such as pandemic, wars, natural calamities etc.