generate_condprob

Define structural function by a conditional probability table

The implemented R6 class 'SCM' aims to simplify working with structural causal models. The missing data mechanism can be defined as a part of the structural model. The class contains methods for 1) defining a structural causal model via functions, text or conditional probability tables, 2) printing basic information on the model, 3) plotting the graph for the model using packages 'igraph' or 'qgraph', 4) simulating data from the model, 5) applying an intervention, 6) checking the identifiability of a query using the R packages 'causaleffect' and 'dosearch', 7) defining the missing data mechanism, 8) simulating incomplete data from the model according to the specified missing data mechanism and 9) checking the identifiability in a missing data problem using the R package 'dosearch'. In addition, there are functions for running experiments and doing counterfactual inference using simulation.

Juha Karvanen

R6causal

R6 Class for Structural Causal Models

generate_condprob function

<dl><dt>ycondx</dt>
<dd>A data table or a data frame with the following structure<ul>
<li>1st column: variable to be generated, "Y"</li>
<li>middle columns: the parents of the the 1st column variable, "X"</li>
<li>last column: the probability the case specified be the other columns, "P(Y|do(X))"</li>
</ul></dd>
<dt>x</dt>
<dd>A data table or a data frame that contains data on the variables in the
middle columns of <code>ycondx</code>, "X" and one or more columns giving data on U-variables.</dd>
<dt>Umerge_expr</dt>
<dd>A character string specifying how the U-variables will
be combined when the value "Y" is generated, e.g. "u" or "(u1+u2)/2". The result of the
expression should be a random number in the interval [0,1].</dd></dl>

Arguments

Define structural function by a conditional probability table — generate_condprob

<dl>

<dt>ycondx</dt>
<dd>A data table or a data frame with the following structure<ul>
<li>1st column: variable to be generated, "Y"</li>
<li>middle columns: the parents of the the 1st column variable, "X"</li>
<li>last column: the probability the case specified be the other columns, "P(Y|do(X))"</li>
</ul></dd>


<dt>x</dt>
<dd>A data table or a data frame that contains data on the variables in the
middle columns of <code>ycondx</code>, "X" and one or more columns giving data on U-variables.</dd>


<dt>Umerge_expr</dt>
<dd>A character string specifying how the U-variables will
be combined when the value "Y" is generated, e.g. "u" or "(u1+u2)/2". The result of the
expression should be a random number in the interval [0,1].</dd>

</dl>

generate_condprob: Define structural function by a conditional probability table

Description

Usage

Value

Arguments

Examples