selBestSubset: Load Estimation
In USGS-R/rloadest: River Load Estimation

Description Usage Arguments Value Note See Also

Select the "best" subset of a user-defined rating-curve (regression) model for rver load estimation.

1
2
3

selBestSubset(formula, min.formula = ~1, data, subset, na.action, flow, dates,
  flow.units = "cfs", conc.units = "", load.units = "kg",
  time.step = "day", station = "", criterion = c("AIC", "SPCC", "AICc"))

`formula`	a formula describing the regression model. See `loadReg` for details.
`min.formula`	a formula containing the minimum variables to use in the final model. The default is to only force the intercept term, whihc will normally be acceptable. In some rare cases, the log of flow may be needed.
`data`	the data to search for the variables in `formula`.
`subset`	an expression to select a subset of the data.
`na.action`	what to do with missing values.
`flow`	character string indicating the name of the flow column.
`dates`	character string indicating the name of the date column.
`flow.units`	character string describing the flow unit.
`conc.units`	character string describing the concentration unit.
`load.units`	character string describing the load unit.
`time.step`	character string describing the time step of the calibration data.
`station`	character string description of the station.
`criterion`	the criterion to use for subset selection, must be one of "AIC," "SPCC," or "AICc."

An object of class "loadReg."

The printed output of the model inlcudes the anova component from step. That table summarizes the step wise selection and the criterion used for each step. The statistics possibly represent a smaller sample size than used for the final model because the step function requires a data set with no missing values. If missing values are found a warning is printed.

censReg, step

USGS-R/rloadest documentation built on Oct. 2, 2020, 5:21 a.m.