findBadParms: Identify parameters inappropriate for use in the OFPE model...

In paulhegedus/OFPE: On-Farm Precision Experiments (OFPE) Data Management and Analysis Tools

Identify parameters inappropriate for use in the OFPE model fitting.

Description

Identifies whether the parameters listed in a data.frame ('parm_df$parms'), meet the criteria to be omitted from the model, making it a 'bad_parm'. The criteria for this is over 30% of data for a given year missing for a parameter or a standard deviation of zero, indicating singularity. These will lead to errors in the model fitting process and subsequent predictions. It is a conservative parameter selection step that promotes the automation of the process.

Usage

findBadParms(parm_df, dat)

Arguments

parm_df	Data.frame with a column for the parameters to check, named 'parms', a column called 'bad_parms' filled with logical responses on whether to omit the parameter from the model fitting process, and columns labeled 'mean' and 'sd' to populate in the 'bad_parm' identification process. Other columns may be included but these columns and specific column names are required.
dat	Data.table with the data to check validity of parameters in. Typically the training data that the model will be fit on.

Value

Updated 'parm_df' table.

paulhegedus/OFPE documentation built on Nov. 23, 2022, 5:09 a.m.