forecast_one_fit: Try to estimate and forecast particular request
In bdemeshev/torro:

Description Usage Arguments Details Value Examples

forecast_one_fit takes one row of fits_long and tries to forecast with corresponding model. To protect against possible errors with complex models try is used.

1 2	forecast_one_fit(fit_row, var_sets, ts_data = torro::rus_macro, redundant = NULL)

`fit_row`	one row from data frame with models to estimate.
`var_sets`	data frame with variable names. Should contain at least two columns: `var_set` name of a set, `variable` names of the variables in the set.
`ts_data`	multivariate time series
`redundant`	vector of column names NOT corresponding to parameters. If NULL then built-in vector is used. See `get_parameter_colnames`.

Does not use parallel computations.

forecasts from estimated model or try-error class

1
2
3

fits_long_toy <- create_fits_long(shifts_toy, models_toy, var_sets_toy, horizons_toy)
fit_row <- fits_long_toy[1, ]
attempt <- forecast_one_fit(fit_row, var_sets)