Python – 统计中的倒数逆高斯分布
scipy.stats.recipinvgauss()是倒数逆高斯连续随机变量。它作为rv_continuous 类的实例继承自泛型方法。它使用特定于此特定发行版的详细信息来完成方法。
参数 :
q : lower and upper tail probability
x : quantiles
loc : [optional]location parameter. Default = 0
scale : [optional]scale parameter. Default = 1
size : [tuple of ints, optional] shape or random variates.
moments : [optional] composed of letters [‘mvsk’]; ‘m’ = mean, ‘v’ = variance, ‘s’ = Fisher’s skew and ‘k’ = Fisher’s kurtosis. (default = ‘mv’).
Results : reciprocal inverse Gaussian continuous random variable
代码#1:创建倒数逆高斯连续随机变量
# importing library
from scipy.stats import recipinvgauss
numargs = recipinvgauss .numargs
a, b = 4.32, 3.18
rv = recipinvgauss (a, b)
print ("RV : \n", rv)
输出 :
RV :
scipy.stats._distn_infrastructure.rv_frozen object at 0x000002A9D843A9C8
代码#2:倒数逆高斯连续变量和概率分布
import numpy as np
quantile = np.arange (0.01, 1, 0.1)
# Random Variates
R = recipinvgauss .rvs(a, b)
print ("Random Variates : \n", R)
# PDF
R = recipinvgauss .pdf(a, b, quantile)
print ("\nProbability Distribution : \n", R)
输出 :
Random Variates :
6.540700180076524
Probability Distribution :
[0.03015471 0.03206632 0.03410829 0.03629051 0.03862377 0.04111981
0.04379146 0.04665275 0.04971902 0.05300712]
代码#3:图形表示。
import numpy as np
import matplotlib.pyplot as plt
distribution = np.linspace(0, np.minimum(rv.dist.b, 3))
print("Distribution : \n", distribution)
plot = plt.plot(distribution, rv.pdf(distribution))
输出 :
Distribution :
[0. 0.04081633 0.08163265 0.12244898 0.16326531 0.20408163
0.24489796 0.28571429 0.32653061 0.36734694 0.40816327 0.44897959
0.48979592 0.53061224 0.57142857 0.6122449 0.65306122 0.69387755
0.73469388 0.7755102 0.81632653 0.85714286 0.89795918 0.93877551
0.97959184 1.02040816 1.06122449 1.10204082 1.14285714 1.18367347
1.2244898 1.26530612 1.30612245 1.34693878 1.3877551 1.42857143
1.46938776 1.51020408 1.55102041 1.59183673 1.63265306 1.67346939
1.71428571 1.75510204 1.79591837 1.83673469 1.87755102 1.91836735
1.95918367 2. ]
代码#4:改变位置参数
import matplotlib.pyplot as plt
import numpy as np
x = np.linspace(0, 5, 100)
# Varying positional arguments
y1 = recipinvgauss .pdf(x, 1, 3, 5)
y2 = recipinvgauss .pdf(x, 1, 4, 4)
plt.plot(x, y1, "*", x, y2, "r--")
输出 :