Another very famous approach to regress on time series data is to regress it with its lag term. This genre of models is referred to as auto-regressive models (AR models). The AR models are very good in capturing trends as the next time values are predicted based on the prior time values. Thus, AR models are very useful in situations where the next forecasted value is a function of the previous time period, such as an increase in average stock price gain due to good company growth; we expect the effect to retain over time and price should keep increasing as a function of time as the trend component.
The auto-regressive model is defined as AR(p), where p refers to the order of the AR component.
The first-order AR model ...
