def test_mgc_sample_non_linear():
mgc = MGC()
# spiral data
X1 = np.array([1.19076532, -4.89949601, -1.04671781, -0.78870398, -4.96541929, -3.47913731, 0.61608779, -1.04636872, 0.01792450, -2.99958358, 1.99093892, 2.02476848, -1.00291164, 3.84247195, -0.97867784, 0.06643890, -2.04553564, 3.95870375, -2.60045257, -1.04668339, -0.92919140, -2.94754180, -0.48332416, 1.26781216, 0.13068389,
0.17426063, 1.51532887, -0.65853037, 1.83957953, -2.47333005, 0.13855163, -3.51847437, -4.94476873, -0.99116531, 0.94505361, 0.02117659, -0.98862767, -1.04662608, -0.47863973, -4.95602679, -2.44573834, -0.57811752, 3.35735796, -2.77334193, -2.66310047, 0.12282333, 1.84287144, 3.17489424, 1.69930914, 3.24924651]).reshape(-1, 1)
Y1 = np.array([-1.40947688, 1.04194197, 0.40771759, -3.16609585, 0.62368044, 3.77762740, 4.09967197, -0.45247854, 0.21346485, -1.51231169, 0.01814165, 0.63590620, -0.17373739, 0.88791421, 0.29489715, -0.25420314, 1.60145057, 0.70182757, -2.24325328, -0.52412790, 3.37903326, -1.25394361, -2.56437048, -1.63174998, 0.40462931,
1.49760321, 1.42923722, 0.23255328, 1.35751771, -2.10603089, 2.85151995, 3.03632266, 0.48248123, 0.54841016, 2.22176329, 0.77264609, 0.41259043, -0.50495140, -1.44860013, 1.20476810, -1.94893291, -1.52795955, -2.06691591, -1.83411979, 1.07099382, 0.24942774, 1.43385633, 2.94960357, -3.34529118, 3.16868061]).reshape(-1, 1)
mgc_statistic_1 = 0.1991505
optimal_scale_1 = [2, 6]
mgc_statistic_res_1, independence_test_metadata_1 = mgc.test_statistic(X1, Y1)
assert np.allclose(mgc_statistic_1, mgc_statistic_res_1)
assert np.allclose(optimal_scale_1, independence_test_metadata_1["optimal_scale"])
X2 = np.array([0.01421334, -2.74656298, 0.77314408, 1.16537193, -2.90519144, -4.22264528, -0.89891050, 3.84718386, -1.56702656, 0.96299667, -4.04908889, 1.23211283, 0.16186615, -2.80339400, -4.52963579, 1.50649684, -2.19419173, -0.17158617, -4.99565629, 4.00691706, -2.84808717, 2.21568866, -0.42059251, -2.72486898, 1.93019558,
0.17617120, 3.81677009, -0.22529392, 1.26280821, 3.86391161, 2.02327480, 1.99630927, 1.82478940, -4.44260943, -0.98931338, -2.94800338, -3.01352134, -2.25180778, -0.92833990, -0.76114888, -0.62939441, -0.98264787, -3.82053926, 1.04258668, -2.15039133, -1.03893152, 1.93208834, 1.82361258, 0.99874067, 0.24155370]).reshape(-1, 1)
Y2 = np.array([0.038971270, 0.833857761, -1.521444608, 4.304165019, -1.020602336, 2.312007878, 0.155703456, 1.380484921, 4.254963435, -1.278597445, 2.701115622, 2.021871587, 0.456433211, 0.829949638, 1.885978521, -0.945712267, -2.407083726, -3.454198852, -0.014086007, 0.145404797, 0.670799013, 3.672285741, 0.602226433, 3.785518506, -0.350152540, -
1.588788123, 1.520267990, -1.409208061, -1.398139886, 1.131541762, 0.356061760, 0.002080002, 1.017373457, 2.080078016, -0.042361065, -0.973710472, -0.228119833, 1.550782393, 0.203302174, -3.383666925, 0.576408069, 0.209023397, 2.955761299, -1.346304143, 4.215744704, -0.540418117, 0.931602249, -0.445806493, 2.179037240, -1.662197212]).reshape(-1, 1)
mgc_statistic_2 = 0.2313806
optimal_scale_2 = [3, 4]
mgc_statistic_res_2, independence_test_metadata_2 = mgc.test_statistic(X2, Y2)
assert np.allclose(mgc_statistic_2, mgc_statistic_res_2)
assert np.allclose(optimal_scale_2, independence_test_metadata_2["optimal_scale"])
def test_mgc_sample_linear():
X = np.array([0.07487683, -0.18073412, 0.37266440, 0.06074847, 0.76899045,
0.51862516, -0.13480764, -0.54368083, -0.73812644, 0.54910974]).reshape(-1, 1)
Y = np.array([-1.31741173, -0.41634224, 2.24021815, 0.88317196, 2.00149312,
1.35857623, -0.06729464, 0.16168344, -0.61048226, 0.41711113]).reshape(-1, 1)
mgc_statistic = 0.4389398
optimal_scale = [10, 10]
mgc = MGC()
mgc_statistic_res, independence_test_metadata = mgc.test_statistic(X, Y)
assert mgc.get_name() == 'mgc'
assert np.allclose(mgc_statistic, mgc_statistic_res)
assert np.allclose(optimal_scale, independence_test_metadata["optimal_scale"])
def compute_mgc(X, Y):
mgc = MGC()
mgc_statistic, independence_test_metadata = mgc.test_statistic(X, Y)
p_value, metadata = mgc.p_value(X, Y)
# mgc_statistic determines the correlation between the variables.
# so, if the variable is closer to -1 or 1, and not extremely close to
# 0, then there might be a correlation. If the p-value (the uncertainty,
# from the way I see it) is lower than 0.05 or 5%, then there might be a
# correlation. Make sure you are running the unbiased version of the method,
# otherwise you might be lead to false results.
print("MGC test statistic:", mgc_statistic)
print("P Value:", p_value)
#print("Optimal Scale:", independence_test_metadata["optimal_scale"])
return mgc_statistic, p_value, independence_test_metadata
def test_mgcx():
# test zero variance dataset with multiple lags.
X = np.array([1, 1, 1, 1, 1, 1, 1])
Y = np.array([1, 2, 3, 4, 5, 6, 7])
mgcx = MGCX(max_lag=3)
assert np.allclose(mgcx.test_statistic(X, Y)[0], 0)
dir_name = './mgcpy/independence_tests/unit_tests/dcorr/data/'
X = np.genfromtxt(dir_name + 'test_stat_X_mtx.csv', delimiter=',')
Y = np.genfromtxt(dir_name + 'test_stat_Y_mtx.csv', delimiter=',')
Y = Y[:, np.newaxis]
mgcx = MGCX(max_lag=0)
mgc = MGC()
# test get_name
assert mgcx.get_name() == 'mgcx'
assert np.allclose(mgcx.test_statistic(X, Y)[0],
mgc.test_statistic(X, Y)[0],
atol=1e-4)
'''
Generate independent random variables and ensure that the test is valid with hypothesis test.
'''
random.seed(123)
mgcx = MGCX(max_lag=2)
n = 20
num_sims = 20
alpha = 0.05
num_rejects = 0
for sim in range(num_sims):
X = np.random.normal(0.0, 1.0, n)
Y = np.random.normal(0.0, 1.0, n)
if mgcx.p_value(X, Y, replication_factor=100)[0] < alpha:
num_rejects += 1
assert np.allclose(num_rejects,
num_sims * alpha,
atol=1.96 * num_sims * alpha * (1 - alpha))
# -*- coding: utf-8 -*-
"""
Created on Fri Feb 15 12:13:46 2019
@author: sandhya
"""
test = 'ACE'
import numpy as np
from mgcpy.independence_tests.mgc import MGC
import matplotlib.pyplot as plt; plt.style.use('classic')
import seaborn as sns; sns.set()
atlases = ('JHU','aal','brodmann','CPAC200','desikan','DK','HarOxCort','HarOxSub','hemispheric','pp264','tissue')
results = np.zeros([1,11])
i = 0
for atlas in atlases:
#atlas = 'tissue'
X = np.load('Xintersect_c_'+atlas+'.npy')
Y = np.load('Yintersect_c_'+atlas+'.npy')
mgc = MGC()
mgc_statistic, independence_test_metadata = mgc.test_statistic(X, Y)
p_value, metadata = mgc.p_value(X, Y)
print(atlas)
print("MGC test statistic:", mgc_statistic)
print("P Value:", p_value)
results[0,i]=p_value
i+=1
np.save(test+'_results',results)