[−][src]Function arima::acf::pacf
pub fn pacf<T: Float + From<u32> + From<f64> + Into<f64> + Copy + AddAssign>(
x: &[T],
max_lag: Option<usize>
) -> Result<Vec<T>, ArimaError>
Calculate the partial auto-correlation coefficients of a time series of length n.
If you already calculated the auto-correlation coefficients (ACF), consider
using pacf_rho
instead.
Arguments
&x
- Reference to input vector slice of length n.max_lag
- Maximum lag to calculate the PACF for. Defaults to n.
Returns
- Output vector of length
max_lag
.
Example
use arima::acf; let x = [1.0, 1.2, 1.4, 1.6]; let pr = acf::pacf(&x, Some(2)).unwrap(); assert!((pr[0] - 0.25).abs() < 1.0e-7); assert!((pr[1] - -0.3866667).abs() < 1.0e-7);