match_observations: Match Observations with Predictions

In kamapu/cropgrowth: Assessing Results from Crop Growth Modelling

Description

Interpolation of predicted values for comparison with observed ones.

This function is using approx() for the interpolation of predicted variables to observations. For the interpolation, the variables indicated in 'match' will be considered as time or progress in the simulated dynamics (e.g. growth).

There are two assumptions that will not be tested by the function. On the one side the predictions are sorted according to the variable 'match'. On the other side, the single values of the variable 'group' represent single runs in the simulation (i.e. single treatments in growth experiments).

For interpolation is carried out in a loop for single values in 'group'. At every run there is a check on occurrence of ties in the variable 'match' at the data frame 'pred' and a check on observations done within the simulated range. If those tests fail, interpolated values will be NAs and a warning will be displayed for the run.

Depending on the variable used for measuring the dynamic progress and depending on its resolution, ties may not be avoidable. In such cases the argument 'clean_ties' can be set as 'clean_ties=TRUE', which will delete all but the last of the tied values in 'pred'.

Usage

match_observations(
  pred,
  obs,
  vars,
  match,
  group,
  suffix = "_pred",
  clean_ties = FALSE,
  ...
)

Arguments

pred, obs	Data frame containing predicted and observed variables, respectively.
vars	Character vector with the name of variables to be compared.
match	Character value indicating the name of the variable used to match observations with predictions. A vector of length 2 can be provided, where the first value is the name of the variable in the data frame 'pred' and the second is the name of the same variable in 'obs'.
group	Character value indicating the name of the variable used for grouping predictions and observations (i.e. simulation runs or treatments). This argument can be also of length 2 as for 'match'.
suffix	Character value used as suffix for predicted variables in output.
clean_ties	Logical value, whether ties in the 'match' variable should be deleted from the data set with predictions or not.
...	Further arguments passed to the function approx.

Value

A copy of the data frame 'obs' including predicted values for the requested variables.

Examples

## No example at the moment

kamapu/cropgrowth documentation built on Aug. 22, 2021, 8:55 a.m.