ARR() defunct in brms: Autoregressive structure for dependant variable?


The ARR() correlation structure is no longer supported in BRMS (‘defunct’). This allowed for the dependant variable to be lagged and made a predictor on the right hand side, y \sim y_{t-1} + y_{t-2} + ... for an ARR(2) process.

What alternative exists to apply this in BRMS now? I can’t see anything on the help pages.



You can specify such a contrast manually as one or more predictor terms which just represent lagged y values.

1 Like

Is this not the same as the ar() function in the current version of brms (2.14.4; e.g., see here)?