Calculate the minimum error to assume in order to pass the variance test

This function finds the smallest relative error still resulting in passing the chi-squared test as defined in the FOCUS kinetics report from 2006.

mkinerrmin(fit, alpha = 0.05)

Arguments

fit

an object of class mkinfit.

alpha

The confidence level chosen for the chi-squared test.

Value

A dataframe with the following components:

err.min

The relative error, expressed as a fraction.

n.optim

The number of optimised parameters attributed to the data series.

df

The number of remaining degrees of freedom for the chi2 error level calculations. Note that mean values are used for the chi2 statistic and therefore every time point with observed values in the series only counts one time.

The dataframe has one row for the total dataset and one further row for each observed state variable in the model.

Details

This function is used internally by summary.mkinfit.

References

FOCUS (2006) “Guidance Document on Estimating Persistence and Degradation Kinetics from Environmental Fate Studies on Pesticides in EU Registration” Report of the FOCUS Work Group on Degradation Kinetics, EC Document Reference Sanco/10058/2005 version 2.0, 434 pp, http://esdac.jrc.ec.europa.eu/projects/degradation-kinetics

Examples

SFO_SFO = mkinmod(parent = mkinsub("SFO", to = "m1"),
                  m1 = mkinsub("SFO"),
                  use_of_ff = "max")
#> Temporary DLL for differentials generated and loaded

fit_FOCUS_D = mkinfit(SFO_SFO, FOCUS_2006_D, quiet = TRUE)
#> Warning: Observations with value of zero were removed from the data
round(mkinerrmin(fit_FOCUS_D), 4)
#>          err.min n.optim df
#> All data  0.0640       4 15
#> parent    0.0646       2  7
#> m1        0.0469       2  8
# \dontrun{
  fit_FOCUS_E = mkinfit(SFO_SFO, FOCUS_2006_E, quiet = TRUE)
  round(mkinerrmin(fit_FOCUS_E), 4)
#>          err.min n.optim df
#> All data  0.1544       4 13
#> parent    0.1659       2  7
#> m1        0.1095       2  6
# }