I'd like to analyse repeated observations of persons using type = twolevel. Observations are assessed at the within-group and persons are assessed at the between-group level. The data structure is in a long format. This makes me wonder how one might check whether autocorrelation is present in the data. Is testing for autocorrelation possible at all (sorry for this general question) once the data are stacked in a univariate vector? If so, which syntax would you recommend?
What you are asking about is covered by the new time series features of Mplus Version 8. See UG examples 9.30 and on. See also our new workshops videos on our home page from the August 17-18 workshop at the Johns Hopkins university.