摘要：
importnumpyasnpimportscipy.integrateasintegrateaverage_,standard_deviation=0,1f=lambdax:(np.exp(-(x-average_)**2/(2*standard_deviation**2)))/(np.sqrt(2*np.pi))/standard_deviationr=integrate.quad(f,-1.
import numpy as np
import scipy.integrate as integrate
average_, standard_deviation = 0, 1
f = lambda x: (np.exp(-(x - average_) ** 2 / (2 * standard_deviation ** 2))) / (
np.sqrt(2 * np.pi)) / standard_deviation
r = integrate.quad(f, -1.96, 1.96)
x, step, l = 1, 0.01, []
for i in range(200):
x += step
r_ = integrate.quad(f, -x, x)
l.append([x, r_[0], r_[1]])
l_6 = []
while r_[0] > 0.6:
x -= step
r_ = integrate.quad(f, -x, x)
l_6.append([x, r_[0], r_[1]])
l_6 = sorted(l_6, reverse=False, key=lambda l: l[0])
n99, l_99 = float('{}{}'.format('0.', ''.join(['9' for i in range(9)]))), []
while r_[0] < n99:
x += step
r_ = integrate.quad(f, -x, x)
l_99.append([x, r_[0], r_[1]])
l_99 = sorted(l_99, reverse=True)
dd = 9
90% 1.65
<class 'list'>: [1.6400000000000006, 0.8989948330517928, 6.786100643460162e-14]
<class 'list'>: [1.6500000000000006, 0.9010570639327039, 8.072601275034688e-14]
95% 1.96
<class 'list'>: [1.9500000000000008, 0.9488238809567229, 9.078658007177475e-12]
<class 'list'>: [1.9600000000000009, 0.9500042097035593, 1.047408678736607e-11]
6个9 4.90
<class 'list'>: [4.889999999999939, 0.9999989916402233, 7.499501297747457e-10]
<class 'list'>: [4.899999999999939, 0.9999990416334469, 7.649097605311184e-10]
1.01 68.75%
