Likelihood and Bayesian updating
Likelihood is a function of a parameter given observed data.
Key knowledge
For iid observations, L(θ;x)=∏ᵢ f(xᵢ|θ). Maximum likelihood maximizes this function. With a prior density π, the posterior is proportional to L(θ;x)π(θ), normalized to integrate to one.
Conditions and scope
- State the sampling model and parameter space; the product assumes conditional independence.
- Use a proper prior and a finite positive normalizing integral; no universal optimality of an estimator is asserted.
Sources · 2
References
Sources retain their recorded scope. A source link alone does not establish verification of every statement.
Zhou Fan / Stanford University2 sources
Evidence and review
Section 13.1: iid likelihood and maximum-likelihood estimation. Section 20.1, physical pages 1–2: conditional PMF/density, prior, likelihood and posterior normalization; equations 20.1–20.2.
2026-10-10 · Iid likelihood and proper-prior posterior, continuous parameter form.
Entry details
Permanent link to this entry ↗- Record identity
- math-likelihood-bayes-inference
- Revision
- university-reference.1
- Structured reference record
- University reference