Primers • Autoregressive Models
What is an Autoregressive Model?
 An autoregressive model learns from a series of timed steps and takes measurements from previous actions as inputs for a regression model, in order to predict the value of the next time step.
How does an Autoregressive Model Work?

Autoregression modeling centers on measuring the correlation between observations at previous time steps (the lag variables) to predict the value of the next time step (the output).

If both variables change in the same direction, for example increasing or decreasing together, then there is a positive correlation. If the variables move in opposite directions as values change, for example one increasing while the other decreases, then this is called negative correlation. Either way, using basic statistics, the correlation between the output and previous variable can be quantified.

The higher this correlation, positive or negative, the more likely that the past will predict the future. Or in machine learning terms, the higher this value will be weighted during deep learning training.

Since this correlation is between the variable and itself at previous time steps, it is referred to as an autocorrelation.

In addition, if every variable shows little to no correlation with the output variable, then its likely that the time series dataset may not be predictable.
Citation
If you found our work useful, please cite it as:
@article{Chadha2020DistilledAutoregressiveModels,
title = {Autoregressive Models},
author = {Chadha, Aman},
journal = {Distilled AI},
year = {2020},
note = {\url{https://aman.ai}}
}