Autoregressive Integrated Moving Average - Forecasts Using ARIMA Models

Forecasts Using ARIMA Models

ARIMA models are used for observable non-stationary processes that have some clearly identifiable trends:

  • a constant trend (i.e. zero average) is modeled by
  • a linear trend (i.e. linear growth behavior) is modeled by
  • a quadratic trend (i.e. quadratic growth behavior) is modeled by

In these cases the ARIMA model can be viewed as a "cascade" of two models. The first is non-stationary:

Y_t = \left( 1-L \right)^d X_t

while the second is wide-sense stationary:

\left( 1 - \sum_{i=1}^p \phi_i L^i \right) Y_t = \left( 1 + \sum_{i=1}^q \theta_i L^i \right) \varepsilon_t \, .

Now standard forecasts techniques can be formulated for the process, and then (having the sufficient number of initial conditions) can be forecast via opportune integration steps.

Read more about this topic:  Autoregressive Integrated Moving Average

Famous quotes containing the word models:

    “French rhetorical models are too narrow for the English tradition. Most pernicious of French imports is the notion that there is no person behind a text. Is there anything more affected, aggressive, and relentlessly concrete than a Parisan intellectual behind his/her turgid text? The Parisian is a provincial when he pretends to speak for the universe.”
    —Camille Paglia (b. 1947)