Does anyone know how to make a pretty visualization of the PDF of multivariate (bivariate for simply) normal distribution, with each variable's distribution is projected, like the below figure? Thanks in advance.
Source of the figure:
from this thesis.
This plot is almost certainly produced using matplotlib. Take a look at their tutorials. Stack Overflow also has a matplotlib tag.
To plot in 3D you need to use the mplot3d toolkit.
try this script
from mpl_toolkits.mplot3d import axes3d
import matplotlib.pyplot as plt
from matplotlib import cm
import numpy as np
def gauss1(x):
return np.exp(-(x**2))
def gauss(x, y):
return gauss1(x)*gauss1(2*y)
fig = plt.figure()
ax = fig.gca(projection='3d')
x = np.linspace(-3, 3, 100)
y = np.linspace(-3, 3, 100)
X, Y = np.meshgrid(x, y)
Z = gauss(X, Y)
ax.plot_surface(X, Y, Z, rstride=2, cstride=2, alpha=0.4, cmap=cm.coolwarm)
cset = ax.contourf(X, Y, Z, zdir='x', offset=-4, cmap=cm.coolwarm)
cset = ax.contourf(X, Y, Z, zdir='y', offset=4, cmap=cm.coolwarm)
ax.set_xlabel('X')
ax.set_ylabel('Y')
ax.set_zlabel('Z')
plt.show()
This produces the following result
I am not sure how to plot only the contour lines for the projections, but this should get you started.
