I am trying to produce forecasts using regression models with ARIMA errors, but they always fail to produce anything other than a NULL model; in contrast, TSLM models work just fine on the same data.
Looking for an answer, I found this question on applying models to multiple time-series and tried to reproduce Rob Hyndman's example (copy-paste the code in rstudio cloud).
It didn't work (details below).
What's wrong?
library(tidyverse)
library(tsibble)
library(fable)
library(lubridate)
set.seed(1)
ar1 <- arima.sim(model=list(ar=.6), n=30)
ma1 <- arima.sim(model=list(ma=0.4), n=30)
Date <- ymd(paste0("2019-01-",1:30))
tb <- bind_cols(Date=Date, ar1=ar1, ma1=ma1) %>%
gather("Series", "value", -Date) %>%
as_tsibble(index=Date, key=Series)
tb
tb_all <- tb %>% model(arima = ARIMA(value))
tb_all
# A mable: 2 x 2
# Key: Series [2]
Series arima
<chr> <model>
1 ar1 <NULL model>
2 ma1 <NULL model>
R version 4.0.0 (2020-04-24)
Platform: x86_64-pc-linux-gnu (64-bit)
Running under: Ubuntu 16.04.6 LTS
Matrix products: default
BLAS: /usr/lib/atlas-base/atlas/libblas.so.3.0
LAPACK: /usr/lib/atlas-base/atlas/liblapack.so.3.0
locale:
[1] LC_CTYPE=C.UTF-8 LC_NUMERIC=C LC_TIME=C.UTF-8
[4] LC_COLLATE=C.UTF-8 LC_MONETARY=C.UTF-8 LC_MESSAGES=C.UTF-8
[7] LC_PAPER=C.UTF-8 LC_NAME=C LC_ADDRESS=C
[10] LC_TELEPHONE=C LC_MEASUREMENT=C.UTF-8 LC_IDENTIFICATION=C
attached base packages:
[1] stats graphics grDevices utils datasets methods base
other attached packages:
[1] lubridate_1.7.8 gridtext_0.1.1 ggtext_0.1.0 forcats_0.5.0 stringr_1.4.0
[6] dplyr_0.8.5 purrr_0.3.4 tidyr_1.1.0 tibble_3.0.1 ggplot2_3.3.0
[11] tidyverse_1.3.0 patchwork_1.0.0 zoo_1.8-8 tsibble_0.8.6 feasts_0.1.3
[16] fable_0.2.0 fabletools_0.1.3 janitor_2.0.1 eurostat_3.6.1 scales_1.1.1
[21] readr_1.3.1 here_0.1 devtools_2.3.0 usethis_1.6.1 pacman_0.5.1
loaded via a namespace (and not attached):
[1] nlme_3.1-147 fs_1.4.1 sf_0.9-3 RColorBrewer_1.1-2
[5] httr_1.4.1 rprojroot_1.3-2 tools_4.0.0 backports_1.1.7
[9] utf8_1.1.4 R6_2.4.1 KernSmooth_2.23-16 DBI_1.1.0
[13] colorspace_1.4-1 withr_2.2.0 sp_1.4-2 tidyselect_1.1.0
[17] prettyunits_1.1.1 processx_3.4.2 curl_4.3 compiler_4.0.0
[21] rvest_0.3.5 cli_2.0.2 xml2_1.3.2 desc_1.2.0
[25] classInt_0.4-3 callr_3.4.3 digest_0.6.25 pkgconfig_2.0.3
[29] bibtex_0.4.2.2 sessioninfo_1.1.1 dbplyr_1.4.4 rlang_0.4.6
[33] readxl_1.3.1 rstudioapi_0.11 generics_0.0.2 jsonlite_1.6.1
[37] magrittr_1.5 Rcpp_1.0.4.6 munsell_0.5.0 fansi_0.4.1
[41] RefManageR_1.2.12 lifecycle_0.2.0 stringi_1.4.6 snakecase_0.11.0
[45] pkgbuild_1.0.8 plyr_1.8.6 grid_4.0.0 blob_1.2.1
[49] slider_0.1.3 crayon_1.3.4 lattice_0.20-41 haven_2.3.0
[53] hms_0.5.3 knitr_1.28 anytime_0.3.7 ps_1.3.3
[57] pillar_1.4.4 pkgload_1.0.2 reprex_0.3.0 glue_1.4.1
[61] remotes_2.1.1 modelr_0.1.8 vctrs_0.3.0 cellranger_1.1.0
[65] testthat_2.3.2 gtable_0.3.0 assertthat_0.2.1 xfun_0.14
[69] broom_0.5.6 countrycode_1.2.0 e1071_1.7-3 class_7.3-16
[73] warp_0.1.0 memoise_1.1.0 units_0.6-6 ellipsis_0.3.1
manually installed package urca