version 1.1-2

ChangeLog

@@ -1,3 +1,62 @@
+2010-10-08  Martin Maechler  <[email protected]>
+
+	* R/pcalg.R (pdsep): biCC: speedup
+
+2010-10-07  Markus Kalisch  <[email protected]>
+
+	* R/pcalg.R (fci): Added option "labels" to give nodenames to PAG
+	(myfun): Undid MMs change in plot-fci, since the edge marks were
+	not plotted properly anymore
+
+2010-09-28  Markus Kalisch  <[email protected]>
+
+	* man/gmI.Rd: description & details added
+
+	* man/gmG.Rd: description & details added
+
+	* man/gmD.Rd: description & details added
+
+	* man/gmB.Rd: description & details added
+
+2010-09-10  Martin Maechler  <[email protected]>
+
+	* R/pcalg.R (lm.cov): solve(A, b),  *NOT* solve(A) %*% b  !!!!!!!!
+	(legal.psep): speed up
+	* R/pcalg.R (plot(<fciAlgo>): make 'main' working, as for plot(<pc>)
+
+2010-09-09  Markus Kalisch  <[email protected]>
+
+	* inst/doc/pcalgDoc.Snw (effect): Adapted text and code to fit to
+	the new data files
+
+	* man/skeleton.Rd: Adapted examples to new form of data files
+
+	* man/pc.Rd: Adapted examples to new form of data files
+
+2010-08-11  Markus Kalisch  <[email protected]>
+
+	* tests/test_fci.R: Tests for conservative FCI added
+
+	* tests/test_pc.R: Tests for conservative PC added
+
+2010-08-09  Markus Kalisch  <[email protected]>
+
+	* R/pcalg.R (upd): improved by diego
+	(ucp): improved by diego
+	(discr.path): improved by diego
+	(faith.check): new by diego
+	(checkTriple): new by markus and diego
+	(pc.cons.intern): new by diego
+	(find.upd): improved by diego
+	(triple2numb): new by diego
+	(pc): option 'conservative' included
+	(udag2pdagRelaxed): option 'unfVect' added for use in
+	conservative PC
+	(fci): options 'conservative', 'biCC' and 'cons.rules' added
+	(pdsep): options 'unfVect' and 'biCC' added
+	(udag2pag): option 'unfVect' added for use in
+	conservative PC
+
 2010-04-14  Markus Kalisch  <[email protected]>
 
 	* R/pcalg.R (randomDAG): Bug fix (Empty graph could not be

DESCRIPTION

@@ -1,6 +1,6 @@
 Package: pcalg
-Version: 1.0-2
-Date: 2010-03-26
+Version: 1.1-2
+Date: 2010-10-08
 Author: Markus Kalisch, Martin Maechler, Diego Colombo
 Maintainer: Markus Kalisch <[email protected]>
 Title: Estimation of CPDAG/PAG and causal inference using the IDA
@@ -13,11 +13,13 @@ Description: Standard and robust estimation of the equivalence class of
         selection variables are assumed to be present. Functions for
         causal inference using the IDA algorithm (based on do-calculus
         of Judea Pearl) are provided for CPDAGs.
-Depends: abind, corpcor, methods, MASS, graph, ggm, vcd, RBGL, sfsmisc,
-        robustbase
-Suggests: Rgraphviz
+Depends: methods, abind, corpcor, sfsmisc
+Imports: graph, RBGL, ggm, robustbase, graphics, vcd
+Suggests: MASS, graph, Rgraphviz, ggm
 License: GPL (>= 2)
 URL: http://pcalg.r-forge.r-project.org/
-Packaged: 2010-04-16 07:06:13 UTC; kalisch
 Repository: CRAN
-Date/Publication: 2010-04-16 09:51:38
+Repository/R-Forge/Project: pcalg
+Repository/R-Forge/Revision: 43
+Date/Publication: 2010-10-27 05:09:28
+Packaged: 2010-10-18 21:09:29 UTC; rforge

MM_Notes.org

@@ -0,0 +1,87 @@
+
+* Data: Mit Markus gesprochen (3.Sept.2010):
+
+ - Die data/test_*.rda Daten welche nur in tests/
+   gebraucht werden, sollen nicht "exponiert" werden.
+
+ - Die andern sind momentan "falsch" dokumentiert, sowohl formal (-> R CMD check),
+   als auch inhaltlich {die *Namen* strings sind "dokumentiert"; sonst kein Inhalt}.
+
+ - Auch hat es z.T. mehrere Objekte pro *.rda  die inhaltlich zusammengehören;
+   MM denkt, die sollten wohl zusammen in eine Liste (mit kurzem Namen!).
+
+ Wo gebraucht:
+
+  pcalg$ grep-r binaryData
+  ./man/pc.Rd:data(binaryData)
+  ./man/skeleton.Rd:data(binaryData)
+  pcalg$ grep-r discreteData
+  ./inst/doc/pcalgDoc.Snw:data set \code{discreteData} (which consists of $p=5$ discrete
+  ./inst/doc/pcalgDoc.Snw:data(discreteData)
+  ./man/pc.Rd:data(discreteData)
+  ./man/skeleton.Rd:data(discreteData)
+  pcalg$ grep-r gaussianData
+  ./inst/doc/pcalgDoc.Snw: data(gaussianData)
+  ./inst/doc/pcalgDoc.Snw:\code{gaussianData} (which consists of $p=8$ variables and $n=5000$
+  ./inst/doc/pcalgDoc.Snw:data(gaussianData)
+  ./inst/doc/pcalgDoc.Snw:data(gaussianData)
+  ./inst/doc/pcalgDoc.Snw:data(gaussianData)
+  ./inst/doc/pcalgDoc.Snw:data(gaussianData) ## contains data matrix datG
+  pcalg$ grep-r idaData
+  ./inst/doc/pcalgDoc.Snw:data(idaData)
+  pcalg$ grep-r latentVarGraph
+  ./inst/doc/pcalgDoc.Snw:data(latentVarGraph)
+
+## Here, you get their contents:
+
+ for(f in list.files("~/R/Pkgs/pcalg/data/", full.names=TRUE)) {
+    bf <- basename(f)
+    nam <- sub("\\.rda$", '', bf)
+    cat("\n", bf, ":\n-----------------\n", sep="")
+    attach(f)
+    print(ls.str(pos=2))
+    detach()
+ }
+
+
+* DESCRIPTION:
+** Depends:  too many; I do not believe they are all "dependently" needed.
+
+*** DONE Look at ./NAMESPACE  and replace *ALL*  the  Imports() by
+     ImportsFrom(.)
+
+*** TODO Remove more 'Depends:': vcd;  probably RBGL, abind, corpcor
+    - vcd: I've eliminated it, and all the 'R CMD check' still run
+    - abind: abind() needed in gSquareBin() & *Dis() -- once only, each
+    - corpcor: pseudoinverse() [ -> fast.svd() .. non-optimal "A %% diag(.)" ]
+    - RBGL: .. ok, only a few needed
+    - graph: import quite a few; now found all examples -- is this (not
+      attaching 'graph') acceptable to the current pcalg users ?
+
+
+* Remarks on specific functions and issues:
+
+** gSquareBin(), gSquareDis():
+  - returns a P-value but not the test statistic.  Should *really* return
+    an object of (standard S3 class) "htest" (which contains P-value, test
+    statistic, ...)!
+
+** fci() :
+  - returns the adjacency matrix "@ amat" (among other things),
+    a simple matrix with entries in {0,1,2,3}.
+    It would be nice to allow *sparse* matrices here,
+    e.g., by using the  "Matrix" class from the Matrix package.
+
+    This makes sense mostly if it's realistic to have quite sparse
+    and relatively large sets of variables.
+
+** gAlgo-class: consider using setMethod(., "gAlgo")
+   instead of all methods (plot, summary) for both pcAlgo and fciAlgo
+
+** DONE myCItest() in  der Vignette inst/doc/pcalgDoc.Snw
+   instead of lm() twice, use lm.fit *once* (*multivariate* regression).
+   This will probably be much faster.
+
+** ida():  argument  'all.dags' is never used, i.e., never tested.
+
+** dsepTest(): gibt 0 / 1  zurück; wieso nicht einfach FALSE/TRUE wie dsep()?

NAMESPACE

@@ -1,8 +1,32 @@
-import(MASS, graph, robustbase)
 
+importFrom("robustbase",
+	   covOGK, hard.rejection,
+	   Qn, s_Qn, scaleTau2 #, Sn, s_Sn, s_mad, s_IQR
+	   )
+
+#import(graph)# importFrom(graph, )
+importClassesFrom("graph", graph, graphNEL)
+importMethodsFrom("graph",
+		  ## addEdge,
+		  coerce,# as(., ...)
+		  edges, edgeData, edgeDataDefaults,
+		  nodes,
+		  numNodes, numEdges,
+		  subGraph, ugraph)
+
+importFrom("RBGL",
+	   biConnComp, johnson.all.pairs.sp, separates)
+## also    sp.between -- but only for deprecated beta.special(), hence RBGL::*
+importMethodsFrom("RBGL",
+		  bfs)
+
+
+# importFrom("MASS", mvrnorm)
 importFrom("ggm", isAcyclic,transClos,topOrder)
+
 importFrom("graphics", plot)
 
+## FIXME: do *NOT* export those that are "internal" <--> man/pcalg-internal.Rd
 export(randomDAG,
        wgtMatrix,
        rmvDAG,
@@ -59,10 +83,15 @@ export(randomDAG,
        idaFast,
        legal.psep,
        qreach,
-       plotSG)
+       plotSG,
+       triple2numb,
+       pc.cons.intern,
+       checkTriple,
+       faith.check)
+
+exportClasses("pcAlgo",
+              "fciAlgo",
+              "gAlgo")
 
-exportClasses("pcAlgo")
-exportClasses("fciAlgo")
-exportClasses("gAlgo")
 exportMethods("summary","show","plot")