load_uschange¶

load_uschange(return_array=True)[source]¶

Load US Change forecasting dataset.

An example of a single multivariate time series. The data is the percentage changes in quarterly personal consumption expenditure, personal disposable income, production, savings and the unemployment rate for the US, 1960 to 2016.

This data shows an increasing trend, non-constant (increasing) variance and periodic, seasonal patterns.

Channels: [‘Consumption’, ‘Income’, ‘Production’,: ‘Savings’, ‘Unemployment’]

Series length: 187 Frequency: Quarterly

Parameters:

return_arraybool, default=True: return series as an np.ndarray if True, else as a pd.DataFrame in wide format.

Returns:

np.ndarray or pd.DataFrame: US Change dataset, shape (5,187).

References

[1]

Data for “Forecasting: Principles and Practice” (2nd Edition)

Examples

>>> from aeon.datasets import load_uschange
>>> data = load_uschange()
>>> data.shape
(5, 187)
>>> data = load_uschange(return_array=False)
>>> data.shape
(5, 187)