11 Matplotlib
Plot the function
f ( x ) = sin 2 ( x - 2) e - x 2
over the interval [0; 2]. Add proper axis labels, a title, etc.
Create a data matrix X with 20 observations of 10 variables. Generate a vector b with parameters Then
generate the response vector y = Xb + z where z is a vector with standard normally distributed variables.
Now (by only using y and X), find an estimator for b , by solving
^ b = arg min
b
k Xb - y k 2
Plot the true parameters b and estimated parameters ^b. See Figure 1 for an example plot.
Generate a vector z of 10000 observations from your favorite exotic distribution. Then make a plot that
shows a histogram of z (with 25 bins), along with an estimate for the density, using a Gaussian kernel
density estimator (see scipy.stats). See Figure 2 for an example plot.
11.1:
Code:
import numpy as np
import matplotlib.pyplot as plt
x = np.linspace(0, 2, 100)
y = (np.exp(-(x ** 2))) * (np.sin(x - 2) ** 2)
fig = plt.figure()
ax = fig.add_subplot(1, 1, 1)
ax.plot(x, y)
plt.xlabel('X')
plt.ylabel('Y')
ax.set_title('11.1')
plt.show()
Figure:
11.2:
Code:
import numpy as np
import matplotlib.pyplot as plt
import scipy as sp
from scipy.optimize import leastsq, minimize
def error(b0, X, y):
b0 = np.reshape(b0, (10, 1))
return np.linalg.norm(np.dot(X, b0) - y, ord=2)
n = 20
m = 10
X = np.random.randint(1, 4, (n, m))
X = np.mat(X)
b = np.random.randint(-2, 2,(1, m))
b_vector = (np.mat(b)).T
z = np.random.randn(1, n)
z_vector = (np.mat(z)).T
y_vector = np.dot(X, b_vector) + z_vector
y = np.array(y_vector)
b0 = np.zeros((10, 1))
Para = minimize(error, b0, args=(X, y))
b_ = Para.x
x1 = np.linspace(0, 9, 10)
fig = plt.figure()
p1 = plt.scatter(x1, b, marker='x', c='r')
p2 = plt.scatter(x1, b_, marker='o', c='b')
plt.xlim(0, 10)
plt.ylim(-3, 3)
plt.xlabel('index')
plt.ylabel('value')
plt.legend([p1, p2], ['True coefficents', 'Estimated coefficients'])
plt.show()
Figure:
11.3
Code:
import numpy as np
import matplotlib.pyplot as plt
import seaborn
x = np.random.randn(10000)
plt.hist(x, 25, normed=1, facecolor='b', alpha=1, edgecolor = 'black')
seaborn.kdeplot(x)
plt.show()
Figure: