The high-level interface of tsorder: one call that runs the chosen selector and returns an object carrying the selected order and the whole decision path (criterion values, or the studentised statistic and its threshold), ready for printing and plotting.
Arguments
- x
A numeric vector (a stationary series; see
integration_order()first if in doubt).- method
What to run. Information criteria:
"bic","aic","aicc","hq","fpe","cat","mdl". Fine-Hwang threshold statistics:"pacf"(default),"acf_bartlett","inverse_pacf","jirak","jirak_refit"(Jirak's refitted q-hat(5)),"pacf_robust","acf_robust","long_ar_ols". McMillan degree:"svc","hankel_rank","hankel_svt".- ...
Passed on to the underlying low-level function (
ic_order(),fh_order(),svc_order(),hankel_rank_order(),hankel_svt_order()).
Value
An object of class "tsorder_fit" with elements order,
method, family ("ic", "fh", "refit" or "mcmillan"), path (a tibble
of the decision path) and, for threshold methods, tau.
Examples
set.seed(1)
x <- arima.sim(list(ar = c(1.2, -0.6)), 400)
fit <- order_select(x)
fit
#>
#> ── tsorder fit - pacf ──────────────────────────────────────────────────────────
#> ✔ Selected order (last studentised statistic above the threshold): 2
#> ℹ Threshold: 2.998
#> ℹ n = 400; path has 12 entries - see `autoplot()` or `$path`