stats

Assorted statistical functions.

Descriptive statistics

boxcox(x, lmbda)	Box-Cox transformation of x.
boxcox1p(x, lmbda)	Box-Cox transformation of 1 + x.
boxcox_llf(lam, x)	Log-likelihood for maximum likelihood estimation of the Box-Cox parameters.
boxcox_mle(x[, lam0])	Maximum likelihood estimate of lambda in the Box-Cox transformation.
gmean(x[, weights])	Geometric mean of the values in the sequence x.
hmean(x)	Harmonic mean of the values in the sequence x.
inv_boxcox(y, lmbda)	Inverse with respect to x of boxcox(x, lmbda).
inv_boxcox1p(y, lmbda)	Inverse with respect to x of boxcox1p(x, lmbda).
inv_yeojohnson(x, lmbda)	Inverse Yeo-Johnson transformation.
mean(x[, weights])	Mean of the values in the sequence x.
pmean(x, p)	Power (or generalized) mean of the values in the sequence x.
std(x[, ddof])	Standard deviation of the values in the sequence x.
unique_counts(x)	Unique values and their counts in the sequence x.
var(x[, ddof])	Variance of the values in the sequence x.
variation(x[, ddof])	The variation of x.
yeojohnson(x, lmbda)	Yeo-Johnson transformation of x.
yeojohnson_llf(lam, x)	The log-likelihood function for the Yeo-Johnson transformation.
yeojohnson_mle(x[, lam0])	Compute the maximum likelihood estimate of the Yeo-Johnson parameter.

Statistical tests

anova_oneway(*args)	One-way analysis of variance (ANOVA) test.
chisquare(observed, expected[, sum_rel_tol])	Pearson's chi-square test.
fisher_exact(table[, alternative])	Compute Fisher's exact test on the 2x2 table.
gtest(observed, expected[, sum_rel_tol])	G-test (likelihood-ratio test, maximum likelihood test).
odds_ratio(table[, kind, alternative])	Compute the odds ratio for a 2x2 contingency table.
pearsonr(x, y[, alternative])	Pearson's correlation coefficient.
pearsonr_ci(r, n, alpha[, alternative])	Confidence interval of Pearson's correlation coefficient.