find_optimal_coef

find_optimal_coef#

MDDC.MDDC.find_optimal_coef(contin_table, rep=1000, target_fdr=0.05, grid=0.1, exclude_small_count=True, col_specific_cutoff=True, seed=None)#

Find Adaptive Boxplot Coefficient coef via Grid Search. The algorithm can be found at Algorithms

This function performs a grid search to determine the optimal adaptive boxplot coefficient coef for each column of a contingency table, ensuring that the target false discovery rate (FDR) is met.

Parameters:#

contin_tablenumpy.ndarray

A matrix representing the I x J contingency table.

repint, optional

The number of simulated tables under the assumption of independence between rows and columns. Default is 1000.

target_fdrfloat, optional

The desired level of false discovery rate (FDR). Default is 0.05.

gridfloat, optional

The size of the grid added to the default value of coef = 1.5 as suggested by Tukey. Default is 0.1.

exclude_small_countbool, optional

Whether to exclude cells with counts smaller than or equal to five when computing boxplot statistics. Default is True.

col_specific_cutoffbool, optional

If True, then a single value of the coefficient is returned for the entire dataset, else when False specific values corresponding to each of the columns are returned.

Returns:#

OptimalCoefnamedtuple

A namedtuple with the following elements:

  • ‘coef’: numpy.ndarray

    A numeric array containing the optimal coefficient coef for each column of the input contingency table.

  • ‘FDR’: numpy.ndarray

    A numeric array with the corresponding false discovery rate (FDR) for each column.

Examples:#

>>> # Example using a simulated contingency table
>>> import numpy as np
>>> contin_table = np.random.randint(0, 100, size=(10, 5))
>>> find_optimal_coef(contin_table)
Contents