Skip to contents

Modified Arcsine-Log Calibration Curve

Source: R/mod_alcc.R
mod_alcc.Rd

This function runs the modified arcsine-log calibration curve to estimate critical soil test values (CSTV) following Correndo et al. (2017)

Usage

mod_alcc(
  data = NULL,
  ry,
  stv,
  target,
  confidence = 0.95,
  tidy = TRUE,
  plot = FALSE
)

logLik_alcc(object, ...)

boot_mod_alcc(data, ry, stv, n = 500, target = 90, confidence = 0.95, ...)

Arguments

data

Optional argument to call and object of type data.frame or data.table containing the stv and ry data, Default: NULL

ry

name of the vector containing relative yield values (%) of type numeric.

stv

name of the vector containing soil test values of type numeric.

target

numeric value of relative yield target (e.g. 90 for 90%) to estimate the CSTV.

confidence

numeric value of confidence level (e.g. 0.95 for significance = 0.05)

tidy

logical operator (TRUE/FALSE) to decide the type of return. TRUE returns a tidy data frame or tibble (default), FALSE returns a list.

plot

logical operator (TRUE/FALSE) to decide the type of return. TRUE returns a ggplot, FALSE returns either a list (tidy == FALSE) or a data.frame (tidy == TRUE).

object

the "object" is the output data frame from approx with resid column

...

when running bootstrapped samples, the ... (open arguments) allows to add grouping variable/s (factor or character) Default: NULL

n

sample size for the bootstrapping Default: 500

Value

returns an object of type ggplot if plot = TRUE.

returns an object of class data.frame if tidy = TRUE,

returns an object of class list if tidy = FALSE.

logLik_alcc: AIC on original scale function

boot_mod_alcc: bootstrapping function

Details

See online-documentation for additional details.

Note

For extended reference, we recommend to visit doi:10.7910/DVN/NABA57 and https://github.com/adriancorrendo/modified-ALCC by Adrian Correndo.

References

Correndo et al. (2017). A modification of the arcsine–log calibration curve for analysing soil test value–relative yield relationships. Crop and Pasture Science, 68(3), 297-304. doi:10.1071/CP16444

See also

eval_tidy,defusing-advanced TDist,cor,cor.test,sd, approx bind,filter nest ggplot,aes,geom_point,scale_manual,geom_rug,geom_abline,geom_path,annotate,labs,theme annotate

Examples

# \donttest{
 # Example 1 dataset
 dat <- data.frame("ry" = c(65,80,85,88,90,94,93,96,97,95,98,100,99,99,100),
                   "stv" = c(1,2,3,4,5,6,7,8,9,10,11,12,13,14,15))
 # Run
 fit_example <- mod_alcc(data = dat, ry = ry, stv = stv, target=90, confidence = 0.95)
#> Warning: 7 STV points exceeded two-times (2x) 
#>   the CSTV for 90% of RY. Risk of leverage. You may consider a sensitivity analysis by 
#>   removing extreme points, re-run the mod_alcc(), and check results.
 fit_example
#> # A tibble: 1 × 18
#>       n     r RMSE_alcc AIC_alcc AIC_sma BIC_sma p_value confidence target  CSTV
#>   <int> <dbl>     <dbl>    <dbl>   <dbl>   <dbl>   <dbl>      <dbl>  <dbl> <dbl>
#> 1    15 0.968      1.45     59.6   -23.5   -20.0 3.30e-9       0.95     90  4.48
#> # ℹ 8 more variables: LL <dbl>, UL <dbl>, CSTV90 <dbl>, n.90x2 <int>,
#> #   CSTV100 <dbl>, n.100 <int>, Curve <list>, SMA <list>
# }