Moving average models

The moving average models use dependency between residual errors to forecast values in the next time period. The model helps you adjust for any unpredictable events such as catastrophic events leading to a share market crash leading to share prices falling, which will happen over time and is captured as a moving average process.

The first-order moving average denoted by MA(1) is as follows:

xt = α - θ1Єt-1 + Єt

The second-order moving average denoted by MA(2) is as follows:

xt = α - θ1Єt-1 - θ2Єt-2+ Єt

The qth order moving average denoted by MA(q) is as follows:

xt = α - θ1Єt-1 - θ2Єt-2 - ... - θqЄt-q+ Єt

Here, Єt is the identically independently-distributed error at time t and follows normal distribution N(0,σ2Є) with ...

Get Practical Time Series Analysis now with the O’Reilly learning platform.

O’Reilly members experience live online training, plus books, videos, and digital content from nearly 200 publishers.