depictr.acf_plot#

depictr.acf_plot(x, kind='acf', lags=None, title=None)#

Autocorrelation or partial autocorrelation as a stem plot.

The correlations come from statsmodels (acf/pacf). The approximate 95% significance band, plus or minus 1.96 / sqrt(n), is drawn as a shaded ribbon: stems reaching past it are the candidate lags.

Parameters:

  • x (pandas.Series or array-like) – The series. Any index is ignored here; only the values are used.

  • kind ({"acf", "pacf"}) – Autocorrelation or partial autocorrelation.

  • lags (int, optional) – Number of lags to show. Defaults to min(10 * log10(n), n - 1) for the ACF and a smaller cap for the PACF, which is undefined beyond n // 2.

  • title (str, optional)

Return type:

plotnine.ggplot