scikit_posthocs.test_durbin

scikit_posthocs.test_durbin(data: List | ndarray | DataFrame, y_col: str | int | None = None, group_col: str | int | None = None, block_col: str | int | None = None, block_id_col: str | int | None = None, melted: bool = False, sort: bool = True) → tuple[float, float, int]

Durbin’s test whether k groups (or treatments) in a two-way balanced incomplete block design (BIBD) have identical effects. See references for additional information [1], [2].

Parameters:

  • data (Union[List, np.ndarray, DataFrame]) –

    An array, any object exposing the array interface or a pandas DataFrame with data values.

    If melted argument is set to False (default), a is a typical matrix of block design, i.e. rows are blocks, and columns are groups. In this case, you do not need to specify col arguments.

    If a is an array and melted is set to True, y_col, block_col and group_col must specify the indices of columns containing elements of correspondary type.

    If a is a Pandas DataFrame and melted is set to True, y_col, block_col and group_col must specify columns names (string).

  • y_col (Union[str, int] = None) – Must be specified if a is a melted pandas DataFrame object. Name of the column that contains y data.

  • group_col (Union[str, int] = None) – Must be specified if a is a melted pandas DataFrame object. Name of the column that contains group names.

  • block_col (Union[str, int] = None) – Must be specified if a is a melted pandas DataFrame object. Name of the column that contains block names.

  • block_id_col (Union[str, int] = None) – Must be specified if a is a melted pandas DataFrame object. Name of the column that contains identifiers of block names. In most cases, this is the same as block_col except for those cases when you have multiple instances of the same blocks.

  • melted (bool = False) – Specifies if data are given as melted columns “y”, “blocks”, and “groups”.

  • sort (bool = True) – If True, sort data by block and group columns.

Returns:

P value, statistic, and number of degrees of freedom.

Return type:

tuple[float, float, int]

Notes

Rows with missing measurements, treatment labels, or block identifiers are excluded. Counts use the remaining observations, so wholly missing blocks and treatments are ignored. The retained design must have equal block sizes, equal treatment replication, and equal pairwise concurrence.

References

Examples

>>> import numpy as np
>>> import scikit_posthocs as sp
>>> x = np.array([[31,27,24],[31,28,31],[45,29,46],[21,18,48],[42,36,46],[32,17,40]])
>>> sp.test_durbin(x)