Autocorrelation

Autocorrelation (Statistical Model Validation) (GWU EMSE-271)
Index | Topics (Logical Lectures) | Lectures | Problems | Readings | Nomenclature | Concepts

"In statistics, the autocorrelation of a random process describes the correlation between values of the process at different points in time, as a function of the two times or of the time difference." - [|Wikipeda]

"In regression analysis using time series data, autocorrelation of the residuals ("error terms", in econometrics) is a problem." - [|Wikipedia]

["Since autocorrelation is a specific type of cross-correlation, it maintains all the properties of cross-correlation." - [|Wikipedia]]

One "way to detect autocorrelations is to use the Durbin-Watson statistic (DW)." "A good rule of thumb" is "if DW is close to 2.0, then there is no evidence of autocorrelation. Although there is not exact statistical test, thre are tables that provide bounds to determine whether the evidence is conclusive or not." - Lattin


 * Sources:**
 * Autocorrelation. (2009, November 30). In //Wikipedia, The Free Encyclopedia//. Retrieved 20:19, December 8, 2009, from []
 * Analyzing Multivariate Data, by James Lattin, Douglas Carroll and Green ([|Amazon]), pp 61-62.
 * EMSE 271, Fall 2009