Probability for Data Science: From Coin Flips to A/B Tests
Learn the probability that every data scientist uses, through one running example, a food-delivery app with 10,000 orders and an A/B test. Simulate before you calculate, then cover conditional probability and Bayes, distributions, the central limit theorem, confidence intervals, maximum likelihood, hypothesis tests, Bayesian inference and Markov chains. Every program was run and the output on the page is the output it produced.
Who it is for
Learners who know a little Python and want a solid, practical foundation in probability for data science, analytics and machine learning. No maths beyond school algebra is assumed.
Before you start
- Basic Python (functions, loops, running a script)
- A little pandas helps (reading a CSV, filtering rows); the sessions show every line they use
- School-level algebra; no calculus is needed
What you will be able to do
- Calculate and simulate probabilities with the rules of probability, conditional probability and Bayes' theorem
- Choose and use the right distribution (Binomial, Poisson, Exponential, Normal, log-normal) for real data with scipy
- Explain and apply the law of large numbers and the central limit theorem to put error bars on an average
- Build confidence intervals, including bootstrap intervals, and fit distributions by maximum likelihood
- Run and read hypothesis tests and A/B tests correctly, including power, multiple testing and peeking
- Update beliefs with Bayesian inference and model processes over time with Markov chains