L23-stats-post

# L23-stats-post

This preview shows page 1. Sign up to view the full content.

This is the end of the preview. Sign up to access the rest of the document.

Unformatted text preview: informative enough • An 1 − α conﬁdence interval ˆn ˆn is a (random) interval [Θ−, Θ+], ˆn ˆn P(Θ− ≤ θ ≤ Θ+) ≥ 1 − α, s.t. ∀θ – often α = 0.05, or 0.25, or 0.01 – interpretation is subtle Properties: ˆ • E [ Θn ] = θ Conﬁdence intervals (CIs) (consistency) MSE: σ 2/n • CI in estimation of the mean ˆ Θn = (X1 + · · · + Xn)/n – normal tables: Φ(1.96) = 1 − 0.05/2 P • Sample mean often turns out to also be the ML estimate. E.g., if Xi ∼ N (θ, σ 2), i.i.d. ˆ |Θ n − θ | ≤ 1.96 ≈ 0.95 (CLT) √ σ/ n 1.96 σ 1.96 σ ˆ ˆ P Θn − √ ≤ θ ≤ Θn + √ ≈ 0.95 n n More generally: let z be s.t. Φ(z ) = 1 − α/2 zσ zσ ˆ ˆ P Θn − √ ≤ θ ≤ Θn + √...
View Full Document

## This note was uploaded on 06/04/2013 for the course EE 6.431 taught by Professor Johntsitsiklis during the Spring '10 term at MIT.

Ask a homework question - tutors are online