Lecture 8. Test statistics and multivariate methods

Statistical Data Analysis

Dmitry V. Naumov (JINR)

Draft Outline

• TODO: likelihood ratio and profile likelihood
• TODO: discriminants and multivariate classifiers
• TODO: calibration and validation

Exercises

Exercise

Show that the likelihood for a Poisson count \(n\) with expectation \(\mu\) is \[ L(\mu)=\frac{\mu^n e^{-\mu}}{n!}. \]