Answers are grounded in this paper and cite their sources.
Testing for randomness is studied for an alternative model of multivariate normality with the autocorrelation specified by a single parameter. First-order autoregression and first-order moving averages are special cases of this model. The likelihood derivative method is used to obtain tests. The asymptotic efficiencies are derived of these tests compared with tests based on the first-order sample serial correlation.
No takes yet. Share an insight, caveat, or question.
James D. Knoke (1975) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: