Autocorrelation describes sample or population observations or elements that are related to each other across time, space, or other dimensions. Correlated observations are common but problematic, largely because they violate a basic statistical assumption about many samples: independence across elements. Conventional tests of statistical significance assume simple random sampling, in which not only each element has an equal chance of selection but also each combination of elements has an equal chance of selection; autocorrelation violates this assumption. This entry describes common sources of autocorrelation, the problems it can cause, and selected diagnostics and solutions.

Sources

What is the best predictor of a student's 11th-grade academic performance? His or her 10th-grade grade point average. What is the best predictor of this year's crude divorce rate? Usually last ...

  • Loading...
locked icon

Sign in to access this content

Get a 30 day FREE TRIAL

  • Watch videos from a variety of sources bringing classroom topics to life
  • Read modern, diverse business cases
  • Explore hundreds of books and reference titles