Time series is appropriate when dealing with time series data. Time series data is data
collected over time at equally spaced intervals. Stock market data is one type of times
series data. Some other examples are:
- Yearly sales data
- Quarterly revenue data
- Yearly gross domestic product data
- etc.
Typically with time series data the researcher is trying to forecast the future using the
knowledge of the past. Some examples:
- Predicting the next years sales
- Predicting the next quarters revenue
- Predicting the next hours stock price
- etc.
Time series data consists of 4 main components:
- trend
- Long term direction the time series data is going in, for example of an
upward trend see figure ??.
- seasonal
- A pattern that occurs within the year, year over year. Usually, it is
monthly or quarterly, see figure ??.
- cyclical
- Like seasonal, cyclical is categorized by a pattern that occurs but over a
much longer period of time and may vary in length, see figure ??.
- irregular
- Unpredictable and random, like the error term in a general linear model.
There are various techniques for analyzing time series data. This section covers only one way
of analyzing time series data. This section focuses on regression models for analyzing time
series data.
