我试图在训练数据上使用 TSLM、theta 和 ATA 对模型进行多步预测,但是当我在拟合函数中设置 h >= 2 时,我得到 NA 或错误。下面的示例代码:
model_fit <- tsibbledata::aus_production %>%
model(mod = TSLM(Beer ~ Gas + trend()))
fitted(model_fit, h = 2)
theta_fit <- tsibbledata::aus_production %>%
model(mod = THETA(Beer))
fitted(theta_fit, h = 2)
Sometimes I get the error below:
Error: Problem with `mutate()` input `..1`.
ℹ `..1 = dplyr::across(...)`.
x `new_data.tbl_ts(.data, round(n))` can't proceed with tsibble of unknown interval.
我感谢任何帮助或见解。谢谢。