math_utility.py

import numpy as np
from cmath import rect, phase
from math import radians, degrees
from scipy.stats import pearsonr
from scipy.stats import chi2


def cart2pol(x, y):
    rho = np.sqrt(x**2 + y**2)
    phi = np.arctan2(y, x)
    return rho, phi


def pol2cart(rho, phi):
    x = rho * np.cos(phi)
    y = rho * np.sin(phi)
    return x, y


# source: https://rosettacode.org/wiki/Averages/Mean_angle#Python
def mean_angle(deg):
    return degrees(phase(sum(rect(1, radians(d)) for d in deg)/len(deg)))


def circ_corrcc(alpha, x):
    """Correlation coefficient between one circular and one linear random
    variable.

    Args:
        alpha: vector
            Sample of angles in radians

        x: vector
            Sample of linear random variable

    Returns:
        rho: float
            Correlation coefficient

        pval: float
            p-value

    Code taken from the Circular Statistics Toolbox for Matlab
    By Philipp Berens, 2009
    Python adaptation by Etienne Combrisson
    """
    if len(alpha) is not len(x):
        raise ValueError('The length of alpha and x must be the same')
    n = len(alpha)

    # Compute correlation coefficent for sin and cos independently
    rxs = pearsonr(x, np.sin(alpha))[0]
    rxc = pearsonr(x, np.cos(alpha))[0]
    rcs = pearsonr(np.sin(alpha), np.cos(alpha))[0]

    # Compute angular-linear correlation (equ. 27.47)
    rho = np.sqrt((rxc ** 2 + rxs ** 2 - 2 * rxc * rxs * rcs) / (1 - rcs ** 2));

    # Compute pvalue
    pval = 1 - chi2.cdf(n * rho ** 2, 2);

    return rho, pval


def circ_r(alpha, w=None, d=0, axis=0):
    """Computes mean resultant vector length for circular data.

    Args:
        alpha: array
            Sample of angles in radians

    Kargs:
        w: array, optional, [def: None]
            Number of incidences in case of binned angle data

        d: radians, optional, [def: 0]
            Spacing of bin centers for binned data, if supplied
            correction factor is used to correct for bias in
            estimation of r

        axis: int, optional, [def: 0]
            Compute along this dimension

    Return:
        r: mean resultant length

    Code taken from the Circular Statistics Toolbox for Matlab
    By Philipp Berens, 2009
    Python adaptation by Etienne Combrisson
    """
    #     alpha = np.array(alpha)
    #     if alpha.ndim == 1:
    #         alpha = np.matrix(alpha)
    #         if alpha.shape[0] is not 1:
    #             alpha = alpha

    if w is None:
        w = np.ones(alpha.shape)
    elif (alpha.size is not w.size):
        raise ValueError("Input dimensions do not match")

    # Compute weighted sum of cos and sin of angles:
    r = np.multiply(w, np.exp(1j * alpha)).sum(axis=axis)

    # Obtain length:
    r = np.abs(r) / w.sum(axis=axis)

    # For data with known spacing, apply correction factor to
    # correct for bias in the estimation of r
    if d is not 0:
        c = d / 2 / np.sin(d / 2)
        r = c * r

    return np.array(r)


def circ_rtest(alpha, w=None, d=0):
    """Computes Rayleigh test for non-uniformity of circular data.
    H0: the population is uniformly distributed around the circle
    HA: the populatoin is not distributed uniformly around the circle
    Assumption: the distribution has maximally one mode and the data is
    sampled from a von Mises distribution!

    Args:
        alpha: array
            Sample of angles in radians

    Kargs:
        w: array, optional, [def: None]
            Number of incidences in case of binned angle data

        d: radians, optional, [def: 0]
            Spacing of bin centers for binned data, if supplied
            correction factor is used to correct for bias in
            estimation of r

    Code taken from the Circular Statistics Toolbox for Matlab
    By Philipp Berens, 2009
    Python adaptation by Etienne Combrisson
    """
    alpha = np.array(alpha)
    if alpha.ndim == 1:
        alpha = np.matrix(alpha)
    if alpha.shape[1] > alpha.shape[0]:
        alpha = alpha.T

    if w is None:
        r = circ_r(alpha)
        n = len(alpha)
    else:
        if len(alpha) is not len(w):
            raise ValueError("Input dimensions do not match")
        r = circ_r(alpha, w, d)
        n = w.sum()

    # Compute Rayleigh's
    R = n * r

    # Compute Rayleigh's
    z = (R ** 2) / n

    # Compute p value using approxation in Zar, p. 617
    pval = np.exp(np.sqrt(1 + 4 * n + 4 * (n ** 2 - R ** 2)) - (1 + 2 * n))

    return np.squeeze(pval), np.squeeze(z)