There are multiple questions exist on this area, however I can not use them to solve my question. I have a data sample and I want to create the confidence interval for its curve. Here, I provide a simple example:
import numpy as np
%matplotlib inline
import matplotlib.pyplot as plt
mean, lower, upper = [],[],[]
ci = 0.2
for i in range (20):
a = np.random.rand(100) # this is the output
MEAN = np.mean(a)
mean.append(MEAN)
std = np.std(a)
Upper = MEAN ci*std
Lower = MEAN-ci*std
lower.append(Lower)
upper.append(Upper)
plt.figure(figsize=(20,8))
plt.plot(mean,'-b', label='mean')
plt.plot(upper,'-r', label='upper')
plt.plot(lower,'-g', label='lower')
plt.xlabel("Value", fontsize = 30)
plt.ylabel("Loss", fontsize = 30)
plt.xticks(fontsize= 30)
plt.yticks(fontsize= 30)
plt.legend(loc=4, prop={'size': 30})
In the above example, I drew