Definition
The law of large numbers says that an average of sufficiently regular repeated observations approaches its expected value as the number of observations grows. The exact theorem depends on independence, identical distribution or weaker dependence and moment conditions. It describes asymptotic stabilization, not a finite-sample error bound by itself.
How it is calculated
The sample average X-bar-n converges in probability to the expected value of X under the theorem's assumptions.
| Symbol | Meaning | Unit |
|---|---|---|
| X | random observation | observation units |
| n | number of observations | count |
| E | expectation | observation units |
Interpretation
Averaging many comparable draws tends to wash out some random fluctuations. It does not wash out systematic bias, changing regimes or dependence that violates the theorem. 'Large' has no universal numeric threshold.
Assumptions and conventions
-
The relevant law's dependence and moment conditions hold.
-
The expectation exists and is the intended target.
The average and expectation share X units; n is a count.
Limitations and common misreadings
-
The law does not make a small sample representative automatically.
-
It does not remove model misspecification or nonstationarity.
References
- Larry Wasserman. (2004). All of Statistics: A Concise Course in Statistical Inference. Springer Texts in Statistics. Chapter 5, Convergence of Random Variables. doi:10.1007/978-0-387-21736-9. Source

