About Statbitall
Most statistics tutorials skip the math or skip the context. You either get a textbook derivation with no code, or a scikit-learn tutorial with no explanation of what the function actually computes. Statbitall covers both. Every post starts with the idea in plain language, works through the formulas, implements them in Python, and ends with when to use the method in practice and when to pick something else.
Foundations. Probability, distributions, the central limit theorem, Bayes' theorem. The concepts that make everything else on the site work.
Statistical Tests. Hypothesis testing, t-tests, ANOVA, p-values, A/B testing. How to make decisions from data without fooling yourself.
ML Models. Regression, decision trees, clustering, PCA, ensembles. Machine learning explained through the statistics it's built on.
Business Analytics. Experiment design, churn modeling, forecasting, Simpson's paradox. Where statistical methods meet real business decisions.
AI and Deep Learning. Loss functions, gradient descent, embeddings, attention, LLMs. The statistical foundations of modern AI.
Every post follows the same six-part structure: the underlying idea, historical root, key assumptions, the math, the code, and business application. Nothing gets skipped. If a method has assumptions that break in practice, the post says so. If the math requires a concept from an earlier post, there is a link.
Statbitall is written by Pius Oyedepo. I've spent over 10 years working in data, analytics, and finance across different industries. My background is in statistics (B.Sc. First Class, M.Sc. with Distinction), and I've built predictive models, risk models, and reporting systems using Python, SQL, and most of the usual tools.
I started Statbitall because I kept running into the same problem: tutorials that showed the code but skipped the math, or textbooks that proved the theorem but never connected it to a real decision. I wanted one place that covered both, from the formula to the Python implementation to the business context.
If you want to connect, find me on LinkedIn.