Definition
Independent and identically distributed sampling means each observation has the same probability distribution and the observations do not provide information about one another under the model. It is a convenient benchmark assumption for standard errors, laws of large numbers and central limit theorems. Financial time series often need explicit checks or alternative dependence assumptions.
How it is calculated
Observations X_1 through X_n are iid draws from the same distribution F.
| Symbol | Meaning | Unit |
|---|---|---|
| X | observation | data units |
| n | sample size | count |
| F | common distribution | probability distribution |
Interpretation
Each draw comes from the same urn, and drawing one does not change the probabilities for the next. That simplifies the math substantially. When volatility, regimes or serial dependence matter, using iid formulas without adjustment changes the question.
Assumptions and conventions
-
The common distribution is stable over the sampling window.
-
Observations are independent under the model.
X retains data units; n is a count and F is a distribution.
Limitations and common misreadings
-
Identically distributed does not mean the observed values are identical.
-
Uncorrelated observations need not be fully independent.
References
- Larry Wasserman. (2004). All of Statistics: A Concise Course in Statistical Inference. Springer Texts in Statistics. Chapter 2, Random Variables. doi:10.1007/978-0-387-21736-9. Source

