Law of Large Numbers

The law of large numbers gives conditions under which an average converges to an expected value as observations accumulate. It concerns the generating process and does not guarantee representative data or monotonic convergence.

The law of large numbers describes convergence of sample averages under conditions. For independent, identically distributed observations with a finite absolute mean, the average converges to the expected value in the relevant weak or strong sense. This is a mathematical result about a sampling process, not a guarantee that every growing dataset represents the population of interest.

Convergence need not be monotonic, and a large sample does not force the next outcome to compensate for earlier ones. Dependence, changing distributions, selective sampling, or undefined expectations require different analysis. A single observation can be informative or refute a universal claim even though it cannot establish a stable frequency by itself.

When to use it

When analyzing data; when building predictive models; when evaluating statistical claims; when quantifying uncertainty.

How it can help

Define the expectation and sampling process, check conditions, and distinguish reduced sampling fluctuation from bias, dependence, or structural change.