我有一个简单的层次模型,有很多个人,我有一个正态分布的小样本。这些分布的平均值也服从正态分布。
import numpy as np
n_individuals = 200
points_per_individual = 10
means = np.random.normal(30, 12, n_individuals)
y = np.random.normal(means, 1, (points_per_individual, n_individuals))
我想使用PyMC3来计算样本中的模型参数。
import pymc3 as pm
import matplotlib.pyplot as plt
model = pm.Model()
with model:
model_means = pm.Normal('model_means', mu=35, sd=15)
y_obs = pm.Normal('y_obs', mu=model_means, sd=1, shape=n_individuals, observed=y)
trace = pm.sample(1000)
pm.traceplot(trace[100:], vars=['model_means'])
plt.show()
我原以为model_means
的后部看起来像我最初的分布方式。但它似乎收敛于30
,即手段的平均值。如何从pymc3模型恢复均值的原始标准差(在我的示例中为12)?
发布于 2015-11-13 01:09:13
这个问题是我为PyMC3的概念而奋斗的。
我需要n_individuals
观察到的随机变量来建模y
和n_individual
随机变量来建模means
。它们还需要先验hyper_mean
和hyper_sigma
作为它们的参数。sigmas
是y
标准差的优先项。
import matplotlib.pyplot as plt
model = pm.Model()
with model:
hyper_mean = pm.Normal('hyper_mean', mu=0, sd=100)
hyper_sigma = pm.HalfNormal('hyper_sigma', sd=3)
means = pm.Normal('means', mu=hyper_mean, sd=hyper_sigma, shape=n_individuals)
sigmas = pm.HalfNormal('sigmas', sd=100)
y = pm.Normal('y', mu=means, sd=sigmas, observed=y)
trace = pm.sample(10000)
pm.traceplot(trace[100:], vars=['hyper_mean', 'hyper_sigma', 'means', 'sigmas'])
plt.show()
https://stackoverflow.com/questions/33661064
复制相似问题