Definition

Autoregressive Model

An autoregressive model predicts the next value in a sequence by regressing a signal on its own previous values. Specifically, given a time series of observations, the model estimates the conditional probability distribution P(xtxt1,,x1)P(x_t \mid x_{t-1}, \ldots, x_1) or a key statistic thereof, such as the conditional expectation E[(xtxt1,,x1)]\mathbb{E}[(x_t \mid x_{t-1}, \ldots, x_1)]. A foundational challenge for autoregressive models is that the number of historical inputs xt1,,x1x_{t-1}, \ldots, x_1 grows with tt, causing each training example to have a different number of features. Two principal strategies address this: (1) conditioning on a fixed-length window of au au recent observations, and (2) maintaining a latent summary state hth_t that compresses the entire history.

0

1

Updated 2026-05-13

Contributors are:

Who are from:

Tags

D2L

Dive into Deep Learning @ D2L