smooth.AutoMSARIMA.plot

AutoMSARIMA.plot(which=[1, 2, 4, 6], level=0.95, legend=False, lowess=True, **kwargs)

Diagnostic plots for the fitted ADAM model (R: plot.adam).

Parameters:

• which (int or list of int, optional) –

  Plot type(s) to produce. Default [1, 2, 4, 6].

  1 — Actuals vs Fitted

  2 — Standardised Residuals vs Fitted

  3 — Studentised Residuals vs Fitted

  4 — |Residuals| vs Fitted

  5 — Residuals² vs Fitted

  6 — Q-Q plot (distribution-specific)

  7 — Actuals and Fitted over time

  8 — Standardised Residuals vs Time

  9 — Studentised Residuals vs Time

  10 — ACF of Residuals

  11 — PACF of Residuals

  12 — Model states over time

  13 — |Standardised Residuals| vs Fitted

  14 — Standardised Residuals² vs Fitted

  15 — ACF of Squared Residuals

  16 — PACF of Squared Residuals

• level (float, optional) – Confidence level for bounds and bands. Default 0.95.

• legend (bool, optional) – Show legend on applicable plots (2, 3, 7, 8, 9). Default False.

• lowess (bool, optional) – Add LOWESS smoothing line to scatter plots. Default True.

• **kwargs – Passed to matplotlib (e.g. figsize).

Returns:

Single figure when which has one element, list otherwise.

Return type:

matplotlib.figure.Figure or list[matplotlib.figure.Figure]

Examples

>>> model = ADAM(model="AAA", lags=[1, 12])
>>> model.fit(y)
>>> figs = model.plot()                   # default: which=[1,2,4,6]
>>> fig  = model.plot(which=7)            # single time-series plot
>>> figs = model.plot(which=[10, 11])     # ACF and PACF

---

Parent Class: AutoMSARIMA