Module functions.matrixmixer.significances
Functions
def alpha_two_to_one_tailed(alpha: float)-
corrects alpha level for a one sided t-test by halfing it
Args
alpha:float- two-tailed alpha level
Returns
float- one-tailed alpha level
def bonferroni_correction(n: int, alpha: float)-
Perform Bonferroni correction on p-values.
Parameters: n (int): number of tests performed alpha (float): Desired family-wise error rate.
Returns: corrected_alpha (float): Corrected significance level after Bonferroni correction.
def find_nominal_significances(table: dict, break_object: dict, code: int, break_value_labels: dict, alpha: float)-
finds all significanes of a nominal variable, sets all variables for the dictionaries objects
Args
table:dict- table object to add significances to
break_object:dict- break object to add significances to
code:int- code of the answer
break_value_labels:dict- value labels of the break
alpha:float, optional- significance level. Defaults to 0.05.
def find_ordinal_significances(table: dict, break_object: dict, break_value_labels: dict, alpha: float)-
finds all significanes of a ordinal or scale variable, sets all variables for the dictionaries objects
Args
table:dict- table object to add significances to
break_object:dict- break object to add significances to
break_value_labels:dict- value labels of the break
alpha:float, optional- significance level. Defaults to 0.05.
def manual_ttest_ind(arr1:-
performs manual t-test statistic calculation for two arrays of data
Args
arr1:np.array- array of sample 1
arr2:np.array- array of sample 2
Returns
tuple(t_stat, p_value): returns the t-statistic as well as the p-value
def manual_ztest_prop(p1: float, n1: float, p2: float, n2: float) ‑> Tuple[float, float]-
performs manual z-test statistic calculation for two samples
Args
p1:float- proportion sample 1
n1:float- size sample 1
p2:float- proportion sample 2
n2:float- size sample 2
Returns
tuple(z_stat, p_value): returns the z-statistic as well as the p-value