Let’s see how to calculate the geometric mean in Python. The geometric mean is a measure of central tendency that is useful for dealing with data that has a wide range of values and is often used in finance, biology, and other fields.
Calculating a geometric mean with the statistics module
To calculate the geometric average, we need to import the statistics module.
Luckily, there is a dedicated function in the statistics module to calculate the geometric mean.
import statistics as s x = [1, 5, 7, 8, 43, 6] geometric_mean = s.geometric_mean(x) print("Geometric mean equals: " + str(round(geometric_mean, 2)))
The statistics.geometric_mean function computes the geometric mean of the values in the list x. In this example, the geometric mean is rounded to two decimal places:
Calculating a geometric mean using numpy
For calculating the geometric mean using the NumPy library, follow this approach:
import statistics as s import numpy as np x = [1, 5, 7, 8, 43, 6] geometric_mean = np.exp(np.mean(np.log(x))) print("Geometric mean equals: " + str(round(geometric_mean, 2)))
In this approach, we first compute the natural logarithm of the data points, calculate their mean, and then exponentiate the result to obtain the geometric mean.
The output is as follows:
Geometric mean equals: 6.45
Calculating a geometric mean using scipy
Alternatively, you can use scipy to calculate a geometric mean.
from scipy.stats import gmean x = [1, 5, 7, 8, 43, 6] geometric_mean = gmean(x) print("Geometric mean equals: " + str(round(geometric_mean, 2)))
As you can see, gmean is the function that you can use to calculate the scipy geometric mean.
By using these methods and libraries, you can efficiently calculate the geometric mean in Python, making it a valuable tool for various statistical and data analysis tasks.