variance = 1
sigma = math.sqrt(variance)
x = np.linspace(mu - 4 * sigma, mu + 4 * sigma, 1000)
title_font = {
'fontname': 'Arial',
'size': '14',
'color': 'black',
'weight': 'normal',
'verticalalignment': 'bottom'
}
plt.figure(1)
plt.subplot(231)
md = extras.pdf_mvsk([mu, sigma, 1, 0])
plt.plot(x, md(x), color='k')
plt.title('SKEW > 2', **title_font)
plt.ylim(0, 0.5)
plt.xlim(-4, 4)
plt.xticks([])
plt.yticks([])
plt.show()
plt.subplot(232)
plt.title('-2 < SKEW < 2', **title_font)
md = extras.pdf_mvsk([mu, sigma, 0, 0])
plt.plot(x, md(x), color='k')
plt.ylim(0, 0.5)
plt.xlim(-4, 4)
# -*- coding: utf-8 -*-
"""
Created on Tue Feb 20 15:12:54 2018
@author: 79127
"""
import numpy as np
import statsmodels.sandbox.distributions.extras as extras
import matplotlib.pyplot as plt
import matplotlib
matplotlib.style.use('seaborn')
GC = extras.pdf_mvsk([1, 2, 4, 2])
GC(0.05)
GC?
from scipy.stats import norm
norm.pdf(0.5)
norm.ppf?
x = np.linspace(norm.ppf(0.01),norm.ppf(0.99), 100)
plt.plot(x, norm.pdf(x),'r-', lw=5, alpha=0.6, label='norm pdf')
# delete ax2 from the figure
plt.delaxes(ax2)
# add ax2 to the figure again
plt.subplot(ax2)
v=j-a
n=np.linspace(-3,3,10000)
sorting = []
indexob = []
pdf_sort = gh.pdf_mvsk( [res7.mean()[0],res7.std()[0]**2,res7.skew()[0],res7.kurt()[0]-3])
pdf_indx = gh.pdf_mvsk( [resind.mean()[0],resind.std()[0]**2,resind.skew()[0],resind.kurt()[0]-3])
for ii in n:
sorting.append(pdf_sort(ii))
indexob.append(pdf_indx(ii))
a=res1.std()
b=res2.std()
c=res3.std()
d=res4.std()
e=res5.std()
f=res6.std()
g=res7.std()
h=res8.std()
i=res9.std()
j=res10.std()
from statsmodels.sandbox.distributions.extras import pdf_mvsk
import numpy as np
from parameters_range_Probmap import mu_range, sigma_range, skew_range, kurt_range
from pylab import plot, show
dist_space = np.arange((int(mu_range) - int(round(2) * round(sigma_range, 2))),
(int(mu_range) + int(round(2) * round(sigma_range, 2))))
diam_phi = ', '.join(map(str, dist_space))
diam_phi = str("[") + str(diam_phi) + str("]")
npar = str(len(dist_space))
a = (mu_range, sigma_range, skew_range, kurt_range)
mf = pdf_mvsk(a)
probs = np.round(abs(mf(dist_space)), 8)
probs2_1 = np.round(probs / sum(probs), 8)
probs2 = ', '.join(map(str, probs2_1))
probs2 = str("[") + str(probs2) + str("]")
#plot(dist_space, probs2_1)
#show()
with open("sample_TGSD.txt", "w") as f1:
f1.write(npar), f1.write('\n')
f1.write(probs2), f1.write('\n')
f1.write(diam_phi)
f1.close()
print "Grain Sizes = ", npar
STD = np.std(x)
SKEW = stats.skew(x)
KURT = stats.kurtosis(x)
)
return out
# creating distribution
for i in range(15):
d = connorav.MSSKDistribution(mu, std, skew, kurt)
x_rand = d.rvs(N)
print(stats.kurtosis(x_rand))
# instancing distrution
# N = 10000
# ============================================================
# Statsmodels
from statsmodels.sandbox.distributions.extras import pdf_mvsk
import numpy as np
from matplotlib import pyplot as plt
pdf = pdf_mvsk([mu, std, skew, kurt])
x = np.linspace(pdf(1e-4), pdf(1-1e-4))
y = np.array([pdf(xi) for xi in x])
plt.plot(x,y)
plt.show()
def func(x, A): pdf_function = extrastats.pdf_mvsk([mu, sigma, skew, kurtosis]) return A * pdf_function(x)