Data#

Back to User Guide

macroforecast.data is the data entry point for the package. It loads official or user-supplied data, normalizes it to one pandas panel contract, and attaches source metadata. The main output is always a DataBundle or DataSpec. This module does not apply stationarity transforms, outlier rules, imputation, feature engineering, model fitting, or evaluation.

The standard panel is a pandas.DataFrame with a DatetimeIndex named date and one macro series per column. Dataset metadata is kept separately and also mirrored in panel.attrs["macroforecast_metadata"]. mf.data.load_*() returns a DataBundle(panel, metadata). mf.data.spec(...) attaches target, horizon, sample-window, and predictor choices to that panel.

Key Callables#

mf.data.load_fred_md downloads or reads the FRED-MD monthly panel, stores official t-codes in metadata, and returns a DataBundle.

mf.data.spec builds a DataSpec from an already-loaded bundle, recording which target to forecast, which horizons to use, what date range is active, and which predictor columns are included.

import macroforecast as mf

# Load the FRED-MD monthly panel (downloads on first call, then caches).
bundle = mf.data.load_fred_md()

# Attach study-level choices: target, horizons, date range, predictors.
data_spec = mf.data.spec(
    bundle,
    target="INDPRO",
    horizons=[1, 3, 6, 12],
    start="1960-01",
    end="2024-12",
    predictors="all",
)

Executed walkthrough#

Loading the panel and inspecting its shape, span, and first rows:

bundle = mf.data.load_fred_md()
panel = bundle.panel
print(panel.shape)                          # (rows, series)
print(panel.index.min().date(), panel.index.max().date())
print(panel.iloc[:3, :4])

(708, 128)
1959-01-01 2017-12-01
                 RPI  W875RX1  DPCERA3M086SBEA    CMRMTSPLx
date
1959-01-01  2289.932   2151.8           18.191  255861.8850
1959-02-01  2299.790   2160.4           18.380  257783.6485
1959-03-01  2314.456   2176.2           18.555  256866.3717

The panel here is the FRED-MD 2018-01 vintage, so it carries 128 series through 2017-12. Exact dimensions and values depend on the vintage you load. Attaching study choices returns a DataSpec that records the target and horizons:

data_spec = mf.data.spec(bundle, target="INDPRO", horizons=[1, 3, 6, 12])
print(data_spec.target, data_spec.horizons)

INDPRO (1, 3, 6, 12)

Reference#