Statistics tutorials and worked examples
Interpret data, uncertainty and evidence with worked examples, probability calculations and simulations.
Probability and distributions
All on this topicCalculate conditional probabilities and test distribution assumptions.
Bernoulli vs binomial: one outcome or a count of trials
Calculate and simulate a binomial count, relate it to Bernoulli trials and test what fails when outcomes share a common cause.
tutorial · 4 min readConditional probability with a table: change the denominator
Calculate P(flagged | faulty) and P(faulty | flagged) from a small inspection table, then connect them through Bayes’ rule.
tutorial · 4 min read
Information and uncertainty
Interpret entropy and distribution mismatch with explicit calculations.
Entropy, cross entropy and KL divergence on the same distribution
Calculate uncertainty, coding mismatch and KL divergence in nats, verify their identity and inspect a zero-probability failure.
tutorial · 4 min read
Estimation and uncertainty
Use intervals and resampling while keeping their assumptions visible.
Bootstrap confidence intervals: work through every resample
Enumerate a small bootstrap distribution, calculate a percentile interval and explain what its uncertainty does and does not represent.
tutorial · 4 min read
Hypothesis testing
Interpret evidence under an explicit null model.
P-values explained with an exact coin-test calculation
Calculate a two-sided binomial p-value, define what counts as extreme, and separate evidence against a model from effect size and practical importance.
explanation · 4 min read
Reading data and claims
Check plots, comparisons and conclusions against the underlying values.
Misleading statistics: 7 examples and how to spot each one
Seven common ways true numbers mislead: truncated axes, missing denominators, base rates, Simpson's paradox, survivorship bias, cherry-picked windows and correlation sold as cause. With a checklist for reading any claim.
explanation · 11 min read
Learning statistics
Build the foundations for evaluating models and experiments.
Statistics and probability for data science and machine learning: what to learn, in order
The statistics and probability a data scientist or machine learning engineer uses, in a seven-step order: what each step covers, a worked example for each, and what to skip.
learning path · 6 min read