The Durbin-Watson test is a statistical procedure for detecting first-order autocorrelation in the residuals of an ordinary least squares regression. It computes the statistic d = Σ(eₜ − eₜ₋₁)² / Σeₜ², where eₜ denotes the residual at observation t. The statistic ranges from 0 to 4: values near 2 indicate no autocorrelation, values below 2 suggest positive autocorrelation, and values above 2 suggest negative autocorrelation. The test compares d against critical values from the Durbin-Watson distribution, which depend on sample size, number of predictors, and significance level. Persistence mechanism: the test endures as a standardized diagnostic routine embedded in statistical software packages (R, Python statsmodels, SAS, Stata), transmitted through econometrics and applied regression textbooks, and applied as a standard step in regression assumption validation workflows. [formal: durbin-watson_test | substrate: behavior | horizon: a moment | explicit: yes | epoch: 0.01]
Accepted ontology entry
durbin watson test
The Durbin-Watson test is a statistical procedure for detecting first-order autocorrelation in the residuals of an ordinary least squares regression. It computes the statistic d = Σ(eₜ − eₜ₋₁)² / Σeₜ², where eₜ denotes the residual at obse…
Definition
Why it is in scope
A statistical hypothesis test for first-order autocorrelation in the residuals of a regression model, computed as the ratio of the sum of squared consecutive residual differences to the sum of squared residuals. Built to persist as a routine diagnostic step in regression assumption checking.
Names and aliases
- durbin watson testen · CANONICAL
Relations from this entry
- cmskmodwa04zunobpymq3i7ifINSTANCE_OF →
The Durbin-Watson test is a specific statistical test used to detect autocorrelation in regression residuals — a specific kind of regression diagnostic. A competent speaker would call the Durbin-Watson test 'a regression diagnostic'.
- cmsm60g0j00jt1q13798ptex4INSTANCE_OF →
The Durbin-Watson test is a specific statistical test used to detect autocorrelation in regression residuals. It follows the standard test structure: null hypothesis of no autocorrelation, a specific test statistic (DW statistic), known sampling distribution under H0, p-value computation, and decision rule based on significance level. A competent speaker would call it a statistical test.
- cmsdai2d503n23vv3e00xn5baDEPENDS_ON →
The durbin watson test detects autocorrelation in regression residuals. Its operational meaning is tied to regression — remove regression and the test has nothing to operate on. Passes the removal test of Law 8.
Relations to this entry
No accepted relations in this direction.
Record identity
- Created
- Aug 8, 2026, 10:53 PM UTC
- Content hash
- 6ab13315376184bc86b044fd36e466cc2ade00a4218c3f93b4b70d799ac70afb