numpy Simple Linear Regression Using np.linalg.lstsq


We use the same dataset as with polyfit:

npoints = 20
slope = 2
offset = 3
x = np.arange(npoints)
y = slope * x + offset + np.random.normal(size=npoints)

Now, we try to find a solution by minimizing the system of linear equations A b = c by minimizing |c-A b|**2

import matplotlib.pyplot as plt # So we can plot the resulting fit
A = np.vstack([x,np.ones(npoints)]).T
m, c = np.linalg.lstsq(A, y)[0] # Don't care about residuals right now
fig = plt.figure()
ax  = fig.add_subplot(111)
plt.plot(x, y, 'bo', label="Data")
plt.plot(x, m*x+c, 'r--',label="Least Squares")

Note: This example follows the numpy documentation at quite closely.