Here I am again in my journey to learn fable package,
I have some issue in my code according below, maybe I have some misunderstanding about this model, I apologize about that
library(fpp3)
library(imputeTS)
library(tidyverse)
serie <- tibble(data = seq(as.Date("1949-01-01"), length = 24, by = "month"),
value1 = sample(1:24),
value2 = sample(1:24)) %>%
as_tsibble() %>%
fill_gaps(.full = TRUE)
serie<- serie %>%
na_kalman()
fit_nn_train <- serie %>%
model(NNETAR(value1 ~ value2 + lag(value1, 5) + trend() + season() + AR(p = 1), n_nodes = 14, scale_inputs = T))
future <- new_data(serie, 1) %>%
mutate(value2 = mean(data_tsbl_train$value2)) %>%
mutate(value1 = 6)
fc <- forecast(fit_nn,new_data = future)
Erro: Problem with mutate() input NNETAR(value1 ~ value2 + lag(value1, 5) + trend() + season() + AR(p = 1), n_nodes = 14, scale_inputs = T).
x Problem with mutate() input .sim.
x I can't use NNETAR to forecast with missing values near the end of the series.
Input .sim is sim_nnetar(.innov).
Input NNETAR(value1 ~ value2 + lag(value1, 5) + trend() + season() + AR(p = 1), n_nodes = 14, scale_inputs = T) is (function (object, ...) ....
For me this error does not make sense, because there is no missing values in this dataset
Some help again will be highly appreciated,
Best
Lucas