Think Stats: Exploratory Data Analysis in Python

About this Book

Think Stats by Prof. Allen B. Downey teaches probability and statistics through code rather than abstract mathematics. Using real datasets, it enables developers to explore data, test hypotheses, and extract meaningful patterns computationally. The book begins with summary statistics, exploratory data analysis (EDA), and probability mass functions (PMFs). Cumulative Distribution Functions (CDFs) are championed as the superior visual tool for comparing empirical distributions. It covers probability density functions (PDFs), analytical distributions (Pareto, exponential, lognormal), and kernel density estimation. Statistical hypothesis testing, null hypothesis rejection, p-values, and effect size calculations are explained through simulations. Correlation and regression analysis are covered with Ordinary Least Squares (OLS) and multiple linear regression models. Bayesian reasoning, prior updating, and likelihood estimation are demystified with intuitive computational examples. Survival analysis, hazard functions, and time-to-event statistical forecasting are applied to real-world medical data. A perfect guide for software engineers, data analysts, and researchers looking to apply statistical thinking to software engineering.