This implementation contrasts the empirical distribution of a measurement variables against assumed distributions. The approach is adapted from the idea of rootograms (Tukey 1977) which is also applicable for count data (Kleiber and Zeileis 2016).
acc_shape_or_scale(
resp_vars,
dist_col,
guess,
par1,
par2,
end_digits,
label_col,
study_data,
meta_data
)variable the name of the continuous measurement variable
variable attribute the name of the variable attribute in meta_data that provides the expected distribution of a study variable
logical estimate parameters
numeric first parameter of the distribution if applicable
numeric second parameter of the distribution if applicable
logical internal use. check for end digits preferences
variable attribute the name of the column in the metadata with labels of variables
data.frame the data frame that contains the measurements
data.frame the data frame that contains metadata attributes of study data
a list with:
SummaryData: data.frame underlying the plot
SummaryPlot: ggplot2 probability distribution plot
SummaryTable: data.frame with the columns Variables and GRADING
This implementation is restricted to data of type float or integer.
Missing codes are removed from resp_vars (if defined in the metadata)
The user must specify the column of the metadata containing probability distribution (currently only: normal, uniform, gamma)
Parameters of each distribution can be estimated from the data or are specified by the user
A histogram-like plot contrasts the empirical vs. the technical distribution