#### This article provides a sample of our book: “Descriptive statistics for data-driven decision making with Python”

**Author(s): Pratik Shukla, Roberto Iriondo**

Data science and **machine learning** are scientific disciplines that are ruled by programming and mathematics. Nowadays, most corporations globally generate immense amounts of data that can be further analyzed and visualized by experts to understand trends and forecast predictions. For instance, we can only perform accurate data visualization if our data is clear and understandable.

However, organizations’ data is (frequently) too messy to tinker with — therefore, finding structures and important patterns in data is a crucial task for data science. Statistics provides the methods and tools to find hidden structures and patterns in data so that specialists can make predictions from them — making statistics the most fundamental step in the data science and machine learning scope. We need statistics to transform observations into information. In machine learning, we use a variety of **algorithms** for prediction, classification, and clustering. Although, there are many useful libraries available to use that will perform mathematical calculations for us.

Nevertheless, we need to know the math behind each of the algorithms and statistical methods we use because knowing these gives us insights into what we are doing and ultimately find our why behind our data-driven decisions.

This work aims to understand the core concepts that form the base for data science, machine learning, and related analytical fields. Our primary goal is to show our readers how to perform calculations and why we need such a methodology. In this book, we try our best to showcase a few core statistical methods with their theories and code examples with python.

Please note that in some cases, the output of python programs may vary from the outputs we get by applying the theoretical concepts — the reason behind it is that we will be using python libraries to display outputs, and in some cases, the programmers that created such libraries used different logic to create their methods. Consequently, we consider it crucial to understand the core logic of what we explain in the theoretical concepts because once we understand the concept, it is relatively easy to write pseudocode and code for the task at hand.

“The quiet statisticians have changed our world; not by discovering new facts or technical developments, but by changing the ways that we reason, experiment, and form our opinions…” ~ Ian Hacking