magic.karma: Train multitude of models on a univariate time series

In snarf-snarf/karma: Complementary tools for automated time-series machine learning.

Description

Train multitude of models on a univariate time series

Usage

magic.karma(y, model_list = 1:11, stacking = F, test_pct_train = "auto",
  test_type_train = "auto", test_pct_valid = "auto",
  test_type_valid = "auto", xreg = NULL, plot = F, stdout = F)

Arguments

y	A univariate time-series vector; type <numeric> or <ts>.
model_list	List of indexes of training models/algorithms in that order: ms-arima, auto-arima, nms-arima, ld-arima, rw-arima, sbj-arima, smbj-arima, nnetar, ets, bats, tbats; <list>
stacking	Whether to use ensemble learning algorithms or not; <T/F>
test_pct_train	Percentage of train-test split in model training (e.g. 70-30), after the model is trained.
test_type_train	Train-test split type for training, i.e. percentage or fixed window; "percentage": test_pct = 12 will be read as the 12 percent of the length of the series; "window": test_pct = 12 will be read as the 12 last time points (e.g. months) of the series.
test_pct_valid	Percentage of train-test split in model validation (e.g. 70-30), after the model is trained.
test_type_valid	Train-test split type for validation, i.e. percentage or fixed window; "percentage": test_pct = 12 will be read as the 12 percent of the length of the series; "window": test_pct = 12 will be read as the 12 last time points (e.g. months) of the series.
xreg	Optional vector or matrix of exogenous regressors; see documentation for Arima(), package 'forecast'.
plot	Option to depict plots during local search; if TRUE (default), AC and PAC plots are active. <logical>
stdout	Option to output optimisation diagnostics during local search; <logical>

Value

Object of class "karma.fit"; (extends class "Arima" from package 'forecast').

See Also

tseries, forecast

Examples

 kmodels = magic.karma(JohnsonJohnson)
 kmodels[[1]]$fit_obj$aicc
 kmodels[[1]]$cv_obj$mape_in
 kmodels[[1]]$cv_obj$mape_out  

snarf-snarf/karma documentation built on May 24, 2019, 7:19 a.m.