Comments on "On estimating the orders of an autoregressive moving-average process with uncertain observations"