simulatedB = np.random.normal(5, 0.5, size=20) # create 20 fake Beijing prices simulatedS = np.random.normal(5, 0.5, size=10) # create 10 fake Seattle prices simulatedB ## ## , , ## array([5.28680292, 5.672704 5.47515532, 4.5042225 6.10792454, 5.55367062, 5.93738803, 5.53429585, 5.00046181, 4.85310013, 5.49085665, 4.91413966, 5.43145621, 4.6731814 4.76162393, 4.30303043, 5.36229575, 5.64092359, 4.96849 4.76309679]) ## simulatedS , ## array([5.25239282, 5.10422605, 5.49375017, 4.73512711, 5.90741389, 4.37950577, 5.07215299, 4.30363852, 4.95806495, 4.94275738]) ## np.mean (simulatedB) - np.mean (simulatedS) ## 0.19683804281616624 Hint: As stated before, the inverse of 10x is log10 x. However, log10 x log10 y is NOT equivalent to log10(x - y)!