Fit rating curve using loess smoother

Usage

rc_loess(
  discharge,
  stage,
  ...,
  data = NULL,
  degree = 2,
  span = 0.75,
  extrapolate = TRUE,
  variance = var_none()
)

Arguments

discharge: <data-masking> Discharge: a vector, or an expression evaluated in data, such as a column name.
stage: <data-masking> Stage: a vector, or an expression evaluated in data, such as a column name.
...: Must be empty. Present so that every argument after it has to be named in full.
data: Optional data frame in which discharge and stage are evaluated.
degree: Degree of local polynomials (1 or 2).
span: Smoothing parameter, passed to stats::loess().
extrapolate: Single logical; allow extrapolation beyond the observed stage range? Default is TRUE. If FALSE, predictions outside the range of the gaugings are NA. (This sets loess's surface to "direct" or "interpolate"; see stats::predict.loess().)
variance: How the variance of the gaugings about the curve is modelled: var_none() (or "none", the default), var_prop() (or "prop"), or var_spec() with the variances. See variance. Under var_prop() the loess curve is refitted in rounds, like the parametric fits.

Value

An rc_loess object; see rating_curve for its contents. A loess curve has no parameters, so curve_parameters is an empty list.

Examples

fit <- rc_loess(discharge, stage, data = thompson)
predict(fit, new_stage = c(1, 3, 6), conflev = 0.95)
#> # A tibble: 3 × 4
#>   stage   fit ci_lwr ci_upr
#>   <dbl> <dbl>  <dbl>  <dbl>
#> 1     1  250.   228.   273.
#> 2     3  823.   796.   850.
#> 3     6 2216.  2192.  2240.