We discuss posterior sampling for two distinct multivariate generalisations of the univariate autoregressive integrated moving average (ARIMA) model with fractional integration. The existing approach to Bayesian estimation, introduced by Ravishanker & Ray, claims to provide a posterior-sampling algorithm for fractionally integrated vector autoregressive moving averages (FIVARMAs). We show that this algorithm produces posterior draws for vector autoregressive fractionally integrated moving averages (VARFIMAs), a model of independent interest that has not previously received attention in the Bayesian literature.
|Original language||English (US)|
|Number of pages||3|
|Journal||Australian and New Zealand Journal of Statistics|
|State||Published - Mar 2019|
All Science Journal Classification (ASJC) codes
- Statistics and Probability
- Statistics, Probability and Uncertainty