nonlinear_process

hyppo.tools.nonlinear_process(n, lag=1, phi=1, sigma=1)

2 nonlinearly dependent time series simulation.

$X_t$ and $Y_t$ are together a bivariate nonlinear process. Noise follows $\mathcal{N}(0,\sigma )$. With lag (1), this is

$$\begin{array}{r}\left[\begin{array}{c}{X}_t\\ Y_t\end{array}\right]=\left[\begin{array}{c}\varphi {ϵ}_t{Y}_{t-1}\\ {\eta }_t\end{array}\right]\end{array}$$

Parameters

• n (int) -- The number of samples desired by the simulation (>= 3).

• lag (float, default: 1) -- The maximum time lag considered between x and y.

• phi (float, default: 1) -- The AR coefficient.

• sigma (float, default: 1) -- The variance of the noise.

Returns

x,y (ndarray of float) -- Simulated data matrices. x and y have shape (n,) where n is the number of samples.

Examples using hyppo.tools.nonlinear_process

Time-Series Sims

previous

cross_corr_ar

next

ts_sim