SYSTEMA CONSTRUCTUM

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…

ACCEPTED THINGcmskyze2305tynobphmmqsnmp

Definition

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]

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

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

Open a related act record