I have a pandas series X as X1, X2, ... Xn
I normalize the X series to have a new Y series with mean=0 and std=1.
I want to plot the histogram of X with 2 xticks, one is original values and the other represented normalized values.
How could I do that with matplotlib?
Update:
import numpy as np
x = np.random.randint (0,100,1000)
y = (x- np.mean(x))/np.std(x)
Now I want to plot the histogram of y, but also show the original values (x), not only values of y.
Here is an example, with a second scaled top axis:
import numpy as np
import matplotlib.pyplot as plt
# Some data
x = 12 + 3*np.random.randn(1000)
x_normed = (x - np.mean(x))/np.std(x)
# Graph
fig, ax1 = plt.subplots()
ax1.hist(x_normed, bins=20)
x1_lim = np.array(ax1.get_xlim())
x2_lim = x1_lim*np.std(x) + np.mean(x)
ax2 = ax1.twiny()
ax2.set_xlim(x2_lim)
ax1.set_ylabel('count')
ax1.set_xlabel('normed x', color='k')
ax2.set_xlabel('x', color='k');
Doing the other way around is I think better:
import numpy as np
import matplotlib.pyplot as plt
# Some data
x = 12 + 3*np.random.randn(1000)
# Graph
fig, ax1 = plt.subplots()
ax1.hist(x, bins=20)
x1_lim = np.array(ax1.get_xlim())
x2_lim = (x1_lim - np.mean(x))/np.std(x)
ax2 = ax1.twiny()
ax2.set_xlim(x2_lim)
ax1.set_ylabel('count')
ax1.set_xlabel('x', color='k')
ax2.set_xlabel('x normed', color='k');
which gives:
