macrosynergy.visuals.acf#

Functions used to visualize autocorrelation and partial autocorrelation functions.

plot_acf(df, cids, xcat, lags=30, alpha=0.05, freq='D', agg='sum', remove_zero_predictor=False, start=None, end=None, blacklist=None, figsize=(16, 9), title=None, ncol=3, share_x=True, share_y=True, zero=False, auto_ylims=True, return_fig=False, **kwargs)[source]#

Plots a facet grid of autocorrelation functions for a given xcat and multiple cids.

Parameters:#

dfpd.DataFrame: The input DataFrame with columns [‘real_date’, ‘cid’, ‘xcat’, ‘value’].
cidsList[str]: List of cids to plot.
xcatstr: The xcat to filter and plot ACFs for.
lagsUnion[int, Sequence], default=30: Number of lags for ACF calculation. If an integer, the lags from 1 to lags are plotted. If a sequence is provided, the lags are plotted as given.
alphafloat, default=0.05: Significance level for the confidence intervals.
freqstr, default=’D’: Frequency to aggregate the data to before computing ACF. Accepts pandas frequency aliases such as ‘D’ (daily), ‘W’ (weekly), ‘M’ (monthly), etc.
aggstr, default=’sum’: Aggregation function to use when resampling to a lower frequency. Common values: ‘sum’, ‘mean’, ‘last’.
remove_zero_predictorbool, default=False: Remove zeros from the input series.
blacklistdict: cross-sections with date ranges that should be excluded from the data frame. If one cross-section has several blacklist periods append numbers to the cross-section code.
startstr: ISO-8601 formatted date string. Select data from this date onwards. If None, all dates are selected.
endstr: ISO-8601 formatted date string. Select data up to and including this date. If None, all dates are selected.
figsizeTuple[float, float], default=(16,9): Figure size for the plot.
titleOptional[str], default=None: Title for the plot.
ncolint, default=3: Number of columns in the facet grid.
share_xbool, default=True: Share x-axis across all subplots.
share_ybool, default=True: Share y-axis across all subplots.
zerobool, default=False: Include the zero lag in the plot.
auto_ylimsbool, default=True: Automatically set the y-axis limits for each subplot.
kwargsDict: Additional keyword arguments for the plot passed directly to Facetplot.lineplot.

plot_pacf(df, cids, xcat, lags=30, alpha=0.05, freq='D', agg='sum', remove_zero_predictor=False, method='ywm', start=None, end=None, blacklist=None, figsize=(16, 9), title=None, ncol=3, share_x=True, share_y=True, zero=False, auto_ylims=True, return_fig=False, **kwargs)[source]#

Plots a facet grid of partial autocorrelation functions for a given xcat and multiple cids.

Parameters:#

dfpd.DataFrame: The input DataFrame with columns [‘real_date’, ‘cid’, ‘xcat’, ‘value’].
cidsList[str]: List of cids to plot.
xcatstr: The xcat to filter and plot PACFs for.
lagsUnion[int, Sequence], default=30: Number of lags for PACF calculation. If an integer, the lags from 1 to lags are plotted. If a sequence is provided, the lags are plotted as given.
alphafloat, default=0.05: Significance level for the confidence intervals.
freqstr, default=’D’: Frequency to aggregate the data to before computing PACF. Accepts pandas frequency aliases such as ‘D’ (daily), ‘W’ (weekly), ‘M’ (monthly), etc.
aggstr, default=’sum’: Aggregation function to use when resampling to a lower frequency. Common values: ‘sum’, ‘mean’, ‘last’.
remove_zero_predictorbool, default=False: Remove zeros from the input series.
methodstr, default=’ywm’: Method for Statsmodel’s PACF calculation. Must be one of [‘ywm’, ‘ywmle’, ‘yw’, ‘ywadjusted’, ‘ols’, ‘ols-adjusted’].
blacklistdict: cross-sections with date ranges that should be excluded from the data frame. If one cross-section has several blacklist periods append numbers to the cross-section code.
startstr: ISO-8601 formatted date string. Select data from this date onwards. If None, all dates are selected.
endstr: ISO-8601 formatted date string. Select data up to and including this date. If None, all dates are selected.
figsizeTuple[float, float], default=(16,9): Figure size for the plot.
titleOptional[str], default=None: Title for the plot.
ncolint, default=3: Number of columns in the facet grid.
share_xbool, default=True: Share x-axis across all subplots.
share_ybool, default=True: Share y-axis across all subplots.
zerobool, default=False: Include the zero lag in the plot.
auto_ylimsbool, default=True: Automatically set the y-axis limits for each subplot.
kwargsDict: Additional keyword arguments for the plot passed directly to Facetplot.lineplot.