OpenAha
00
All courses
Math · Advanced

Statistical Inference

How do scientists decide whether an effect is real or just luck? Learn how hypothesis tests and p-values work, why testing many things at once guarantees false alarms, why extreme results tend to be followed by ordinary ones, how companies run A/B tests, what a confidence interval really promises, and how Bayesians update their beliefs as data arrive.

7 lessons~35 min0/7 complete
1

Testing Ideas

Is it real, or is it luck?

  1. Hypothesis TestsCould it be chance?
  2. Multiple ComparisonsTest enough, and something will be “significant”
2

Traps and Tools

Regression to the mean and A/B tests.

  1. Regression to the MeanWhy extremes don’t last
  2. A/B TestingExperiments at scale
3

Estimating

Confidence intervals and Bayesian updating.

  1. Confidence IntervalsWhat “95% confident” really means
  2. Bayesian UpdatingLearning a coin’s bias from data
  3. The BootstrapResampling your own data