DAR1

Notedocblock

DAR1(α, dist, add_noise)

Discrete AutoRegressive process of order 1.

The DAR(1) model is a simple discrete-valued time series model defined by: \(X_t = B_t X_{t-1} + (1 - B_t) \epsilon_t\) where:

• \[ B_t \]

  is an i.i.d. Bernoulli random variable with parameter \(\alpha\).

• \[ \epsilon_t \]

  is an independent sequence of random variables (the innovation).

Fields

• α::Float64: The autoregressive parameter (probability of selecting the previous value). Must be in \((0, 1)\).
• dist::DiscreteUnivariateDistribution: The distribution of the innovation term \(\epsilon_t\).
• add_noise::Bool: Flag to add small noise.