VARIMA, logs of variables and measures of forecast accuracy

[newsagent]

I have this data set from the book

 library(fpp3)
 google_stock <- gafa_stock |>  filter(Symbol == "GOOG", year(Date) >= 2015) |>  mutate(day = row_number()) |>
  update_tsibble(index = day, regular = TRUE)

google_2015 <- google_stock |> filter(year(Date) == 2015)

And the we apply this VARIMA model

fit<- google_2015 |> model(VARIMA(vars(log(Open), log(High)) ))

Finally I am trying to calculate forecast measures

fit %>% forecast:: accuracy()

Error in `mutate()`:
 In argument: `fit = map(fit, accuracy, measures = measures, ...)`.
Caused by error in `gather()`:
! Can't select columns that don't exist.
✖ Column `Open` doesn't exist.
Run `rlang::last_trace()` to see where the error occurred.

Am I doing something wrong?

[robjhyndman]

For forecast accuracy, you will need to pass forecasts to the accuracy() function.

library(fpp3)
#> ── Attaching packages ──────────────────────────────────────────── fpp3 1.0.3 ──
#> ✔ tibble      3.3.1     ✔ tsibble     1.2.0
#> ✔ dplyr       1.2.1     ✔ tsibbledata 0.4.1
#> ✔ tidyr       1.3.2     ✔ ggtime      0.2.0
#> ✔ lubridate   1.9.5     ✔ feasts      0.5.0
#> ✔ ggplot2     4.0.3     ✔ fable       0.5.0
#> ── Conflicts ───────────────────────────────────────────────── fpp3_conflicts ──
#> ✖ lubridate::date()    masks base::date()
#> ✖ dplyr::filter()      masks stats::filter()
#> ✖ tsibble::intersect() masks base::intersect()
#> ✖ tsibble::interval()  masks lubridate::interval()
#> ✖ dplyr::lag()         masks stats::lag()
#> ✖ tsibble::setdiff()   masks base::setdiff()
#> ✖ tsibble::union()     masks base::union()
google_stock <- gafa_stock |>
  filter(Symbol == "GOOG", year(Date) >= 2015) |>
  mutate(day = row_number()) |>
  update_tsibble(index = day, regular = TRUE)

google_2015 <- google_stock |> filter(year(Date) == 2015)

fit <- google_2015 |> model(VARIMA(vars(log(Open), log(High))))

fit |>
  forecast(h = 20, simulate = TRUE) |>
  accuracy(google_stock)
#> # A tibble: 2 × 12
#>   .model  Symbol .response .type    ME  RMSE   MAE   MPE  MAPE  MASE RMSSE  ACF1
#>   <chr>   <chr>  <chr>     <chr> <dbl> <dbl> <dbl> <dbl> <dbl> <dbl> <dbl> <dbl>
#> 1 VARIMA… GOOG   High      Test  -38.4  41.2  38.4 -5.30  5.30  6.23  3.79 0.679
#> 2 VARIMA… GOOG   Open      Test  -39.6  43.1  39.6 -5.55  5.55  5.61  3.58 0.616

^{Created on 2026-06-16 with reprex v2.1.1}

Note that I've used simulate = TRUE. This is because transformations of multivariate forecasts are not supported analytically.

[newsagent]

Thank you very much.