residuals.loadReg: Extract Model Residuals
In USGS-R/rloadest: River Load Estimation

Description Usage Arguments Details Value Note See Also

Extract the residuals from the load or concetration regression model. The residuals will be the same unless the log of flow is not an explanatory variable.

1
2
3

## S3 method for class 'loadReg'
residuals(object, type = "working",
  suppress.na.action = FALSE, model = c("load", "concentration"), ...)

`object`	an object of class "loadReg"—output from `loadReg`
`type`	The type of residuals, see Details.
`suppress.na.action`	logical, suppress the effects of the `na.action` in the call to `loadReg` and return only the fitted values corresponding to the fitted data.
`model`	the type of model, must be either "load" or "concentration."
`...`	not used, required for other methods.

The value for type can be any one of the following:

Value	Description
"working"	Residuals with censored residuals replaced by their expected values
"response"	Residuals from the linear predictor
"influence"	An estimate of Cook's D values based on "working" residuals
"leverage"	The hat diagonals
"S-L"	The square-root of the absolute value of the residuals with censored residuals replaced by their expected value

Also, any other value of type for residuals.survreg can be used to obtain those residuals. Note that "working" and "response" are defined in the table above, in keeping with older versions of loadReg.

The residuals from the regression as specified by type.

The residuals from the load regression are the same as those from the concentration regression, so there is no option to distinguish among those models.

loadReg

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