Struct compute::timeseries::AR

pub struct AR {
    pub p: usize,
    pub coeffs: Vec<f64>,
    pub intercept: f64,
}

Implements an autoregressive models.

Fields

p: usize

coeffs: Vec<f64>

intercept: f64

Implementations

impl AR

Create a new autoregressive model of order p.

pub fn new(p: usize) -> Self

pub fn fit(&mut self, data: &[f64]) -> &mut Self

Fit the AR(p) model to the data using the Yule-Walker equations.

pub fn predict_one(&self, data: &[f64]) -> f64

Given some data, predict the value for a single timestep ahead.

pub fn predict(&self, data: &[f64], n: usize) -> Vec<f64>

Predict n values ahead. For forecasts after the first forecast, uses previous forecasts as “data” to create subsequent forecasts.

Trait Implementations

impl Debug for AR

fn fmt(&self, f: &mut Formatter<'_>) -> Result

Formats the value using the given formatter.

impl Display for AR

fn fmt(&self, f: &mut Formatter<'_>) -> Result

Formats the value using the given formatter.