DefineSummariesClass: R6 class for parsing and evaluating user-specified summary...
In tmlenet: Targeted Maximum Likelihood Estimation for Network Data

Description Usage Format Details Methods

This R6 class that inherits from can parse and evaluate (given the input data frame) the summary measures defined by functions def.sW and def.sA. The object of this class is generally instantiated by calling functions def.sA or def.sW. The summary expressions (stored in exprs_list) are evaluated in the environment of the input data.frame. Note that the evaluation results of the summary measures are never stored inside this class, data can be stored only inside DatNet and DatNet.sWsA R6 classes.

1	DefineSummariesClass

An R6Class generator object

type - Type of the summary measure, sW or sA, determined by the calling functions def.sW or def.sA.
exprs_list - Deparsed list of summary expressions (as strings).
new_expr_names - The summary measure names, if none were provided by the user these will be evaluated on the basis of variable names used in the summary expression itself.
sVar.names.map - Named list that maps the user specified summary measure names to the corresponding matrix column names from the summary measure evaluation result.

new(type): Instantiate a new object of class DefineSummariesClass by providing a type, "sW" or "sA".
set.new.exprs(exprs_list): Sets the internal summary measure expressions to the list provided in exprs_list.
add.new.exprs(NewSummaries): Adds new internal summary measure expressions to the existing ones, NewSummaries must be an object of class DefineSummariesClass (to enable Object1 + Object2 syntax).

itemremove.expr(SummaryName)Remove expression by name (for removing duplicate 'nF' expressions for repeated calls with def.sW()+def.sW()).

eval.nodeforms(data.df, netind_cl): Evaluate the expressions one by one, standardize all names according to one naming convention (described in def.sW), cbinding results together into one output matrix. data.df is the input data.frame and netind_cl is the input network stored in an object of class NetIndClass.
df.names(data.df): List of variables in the input data data.df gets assigned to a special variable (ANCHOR_ALLVARNMS_VECTOR_0).