Skip to content

Estimates

Point vs Interval Estimates

Type Definition Example
Point estimate Single-value estimate of a parameter \(\bar{x}\) as estimate of \(\mu\)
Interval estimate Range expressing uncertainty around estimate \([\bar{x} - z\frac{s}{\sqrt{n}},\; \bar{x} + z\frac{s}{\sqrt{n}}]\) (95% CI)

Confidence Interval for a Mean (large sample)

Symbol Meaning
\(\bar{x}\) Sample mean
\(s\) Sample standard deviation
\(n\) Sample size
\(z\) z-score for desired confidence (e.g., 1.96 for 95%)

Formula

\[\bar{x} \pm z \cdot \frac{s}{\sqrt{n}}\]

Estimates — Common Methods

Method When to use Pros Cons
Method of moments Simple parametric estimation Easy to compute Can be inefficient / biased
Maximum likelihood (MLE) Parametric models Asymptotically efficient Requires model; can be complex to compute
Bayesian estimation When prior knowledge exists Incorporates prior; yields full posterior Needs prior specification; may be computationally intensive
Nonparametric (bootstrap) For CI without parametric assumptions Flexible; few assumptions Computationally heavy