（numpy.random.standard_t）如何确定自由度？

Question

Please advise me how to determine degrees of freedom

1. I use the follow code to plot a student-t distribution with a histogram

# distribution fitting
param = t.fit(log_rets) 
# now get theoretical values in our interval      
pdf_t = stats.t.pdf(lnspc, param[0], param[1], param[2])
# plot it      
plt.plot(lnspc, pdf_t, label="student t") 
plt.show()

2. I found the param[3] as the following

(3.1512592116146094, 0.00046753124755725029, 0.0061044503250347361)

3. I would generate a random number with student-t, but don't know how to determine degrees of freedom

retvec = []
for i in range (n_obs):
    retvec.append(np.random.standard_t(**<degrees of freedom>**, size=n_assets))

Please feel free to comment.

Regards, Nelson.

Answer 1

这是一个老问题，但是对于任何阅读它的人来说，params [0]这里是自由度（例如，尾巴大小），因此您可以使用scipy随机变量生成随机值：

roll = stats.t.rvs(df=params[0], loc=params[1], scale=params[2], size=n_assets)