Rel 0.1.0

Author: goedman

README.md

@@ -15,7 +15,7 @@
 
 StructuralCausalModels.jl is part of the StatisticalRethinkingJulia eco system and, for now, containg functionality to analyse directed acyclic graph (DAG) based causal models as described in [StatisticalRethinking](https://xcelab.net/rm/statistical-rethinking/),[Causal Inference in Statistics](http://bcs.wiley.com/he-bcs/Books?action=index&bcsId=10288&itemId=1119186846) and [Cause and Correlation in Biology](https://www.cambridge.org/core/books/cause-and-correlation-in-biology/247799189B31939D24BC0F61FD59E9BB#).
 
-My initial goal for this package is to have a way to apply SCM ideas to the examples in [StatisticalRethinking.jl](https://github.com/StatisticalRethinkingJulia), i.e. a working version of `d_separation()`, `basis_set()` and `adjustment_sets()`.
+My initial goal for this package is to have a way to apply SCM ideas to the examples in [StatisticalRethinking.jl](https://github.com/StatisticalRethinkingJulia), i.e. a working version of `basis_set()`, `d_separation()` and `adjustment_sets()`.
 
 StructuralCausalModels.jl will be registered and, once registered, can be installed using
 `] add StructuralCausalModels`.
@@ -40,7 +40,8 @@ and particularly:
 4. Sadeghi, K. (2011). Stable classes of graphs containing directed acyclic
 graphs, implementation as included in ggm.
 
-The latter two have been used as the basis for the Julia implementations of e.g. `shipley_test()`, `d_separation()` and `ancestral_graph()`.
+The latter two have been used as the basis for the Julia implementations of `basis_set()``d_separation()`, `shipley_test()`, `pcor_test()` and `ancestral_graph`.
+
 
 # References
 

docs/src/introduction.md

@@ -2,10 +2,10 @@
 
 StructuralCausalModels.jl (SCM) is a Julia package to analyse directed acyclic graphs (DAGs) as described in [StatisticalRethinking](https://xcelab.net/rm/statistical-rethinking/), [Causal Inference in Statistics](http://bcs.wiley.com/he-bcs/Books?action=index&bcsId=10288&itemId=1119186846) and [Cause and Correlation in Biology](https://www.cambridge.org/core/books/cause-and-correlation-in-biology/247799189B31939D24BC0F61FD59E9BB#).
 
-My initial goal for this package is to have a way to apply SCM ideas to the examples in [StatisticalRethinking.jl](https://github.com/StatisticalRethinkingJulia), i.e. a working version of `d_separation()`, `adjustment_sets()` and `implied_conditional_independencies()`(still being developed).
+My initial goal for this package is to have a way to apply SCM ideas to the examples in [StatisticalRethinking.jl](https://github.com/StatisticalRethinkingJulia), i.e. a working version of `basis_set()`, `d_separation()`, `pcor_test()` and `adjustment_sets()`.
 
 All three above references are great introductions to the use of causal models to help in understanding confounding in statistical models. 
 
 [StructuralCausalModels.jl](https://github.com/StatisticalRethinkingJulia/StructuralCausalModels.jl) is part of the [StatisticalRethinkingJulia](https://github.com/StatisticalRethinkingJulia) eco system. The package, once registered, can be installed using `] add StructuralCausalModels`.
 
-Please report issues or file pull requests on Github if you find a problem.
+Please report issues (or file a pull request) on Github if you find a problem.

src/StructuralCausalModels.jl

@@ -55,10 +55,6 @@ include("utils/ggm_conversions.jl")
 include("utils/dagitty_conversions.jl")
 include("utils/graphviz_conversions.jl")
 
-include("test_methods/test_ag.jl")
-
-#include("methods/all_edges.jl")
-
 export
   scm_path
 

src/methods/all_edges.jl

@@ -35,8 +35,8 @@ end
 
   include("test_sr6_4_3b.jl")
   include("test_ag.jl")
-  include("test_agb.jl")
-  include("test_dag.jl")
+  include("test_set_dag_df.jl")
+  include("test_dag_formulations.jl")
   include("test_dagitty_conversion.jl")
   include("test_ggm_conversion.jl")
   include("test_graphviz_conversions.jl")

src/not_used/convert2csv.jl

@@ -2,7 +2,7 @@ using StructuralCausalModels, Test
 
 ProjDir = @__DIR__
 
-#include(scm_path("test_methods", "test_ag.jl"))
+include(scm_path("..", "test", "test_methods", "test_ag.jl"))
 
 amat_data = transpose(reshape([
   0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,
@@ -36,12 +36,12 @@ Sys.isapple() && run(`open -a GraphViz.app $(fn)`)
 m = [:n3, :n5, :n6, :n15, :n16];
 c = [:n4, :n7];
 
-fr = StructuralCausalModels.test_ag(a; m=m, c=c)
+fr = test_ag(a; m=m, c=c)
 
 fr1 = ancestral_graph(a; m=m, c=c)
 @test all(fr .== fr1)
 
-fr2 = StructuralCausalModels.test_ag(dag.a; m=m, c=c)
+fr2 = test_ag(dag.a; m=m, c=c)
 @test all(fr .== fr2);
 
 println()

src/not_used/doc_template.jl

@@ -0,0 +1,31 @@
+using StructuralCausalModels, Test
+
+d1 =OrderedDict(
+  [:u, :w] => :v,
+  [:u] => :x,
+  :s1 => [:u],
+  :w => :y,
+  [:s2] => [:w]
+)
+
+e1 = StructuralCausalModels.edge_matrix(d1)
+
+d2 = OrderedDict(
+  :u => [:x, :v],
+  :s1 => [:u],
+  :w => [:v, :y],
+  :s2 => [:w]
+);
+
+e2 = StructuralCausalModels.edge_matrix(d2)
+
+@testset "dag_formulations" begin
+
+  syms = names(e, 1)
+  for s1 in syms
+    for s2 in syms
+      @test e1[s1, s2] == e2[s1, s2]
+    end
+  end
+
+end