Merge pull request #11 from brandmaier/master

Syncing
brandmaier · Apr 27, 2021 · ab9796b · ab9796b
2 parents 4603780 + 5acdf40
commit ab9796b
Show file tree

Hide file tree

Showing 18 changed files with 123 additions and 49 deletions.
diff --git a/.travis.yml b/.travis.yml
@@ -2,3 +2,44 @@
 
 language: R
 cache: packages
+dist: focal
+r:
+    - release
+    - devel
+r_binary_packages:
+    - bitops
+    - sets
+    - digest
+    - rpart
+    - rpart.plot
+    - plotrix
+    - cluster
+    - stringr
+    - lavaan
+    - expm
+    - lavaan
+    - tidyr
+    - viridis
+    - ggplot2
+    - strucchange
+    - sandwich
+    - zoo
+    - crayon
+    - testthat
+    - Rcpp
+    - RcppEigen
+    - MASS
+    - Matrix
+    - StanHeaders
+    - mvtnorm
+    - xfun
+    - mime
+    - RcppParallel
+    - rmarkdown
+    - htmltools
+    - knitr
+    - rpf
+    - BH
+    - OpenMx
+
+
diff --git a/DESCRIPTION b/DESCRIPTION
@@ -48,6 +48,6 @@ License: GPL-3
 Encoding: UTF-8
 LazyLoad: yes
 Version: 0.9.15
-Date: 2020-10-24
-RoxygenNote: 7.1.0
+Date: 2021-04-26
+RoxygenNote: 7.1.1
 VignetteBuilder: knitr
diff --git a/Makefile b/Makefile
@@ -15,7 +15,7 @@ readme:
 
 build:
 	cd ..;\
-	R CMD build --with-keep.source $(PKGSRC)
+	R CMD build $(PKGSRC)
 
 buildfast:
 	cd ..;\
@@ -29,6 +29,9 @@ install:
 check: build
 	cd ..;\
 	R CMD check --as-cran $(PKGNAME)_$(PKGVERS).tar.gz
+
+dockercheck:
+	docker run r-devel-local `R CMD check --as-cran ../$(PKGNAME)_$(PKGVERS).tar.gz`
 
 clean:
 	cd ..;\

diff --git a/NAMESPACE b/NAMESPACE
@@ -47,8 +47,8 @@ export(toTable)
 export(varimp)
 export(varimpConvergencePlot)
 import(OpenMx)
-import(bitops)
 import(rpart)
+importFrom(bitops,bitAnd)
 importFrom(grDevices,heat.colors)
 importFrom(graphics,barplot)
 importFrom(graphics,hist)

diff --git a/NEWS.md b/NEWS.md
@@ -1,29 +1,37 @@
-# semtree 0.9.14
+# semtree 0.9.15 (2021)
+
+- roll-out of score-based tests - use method "score" to enable 
+- use of method "cross-validation" is now deprecated and discouraged
+- added variable importance for focus parameters
+- added new vignettes
+- fixed compatibility with new OpenMx version
+
+# semtree 0.9.14 (2020)
 
 - added requirement rpart.plot > 3.0.6 to avoid error w.r.t. to missing round.int
 - Manuel Arnold joined the team to work on a score-test implementation
 - fixed behavior of min.N in semtree.control()
 - semtree captions uses signif for truncating decision thresholds to three significant digits
 - fixed class type checks to be compatible with upcoming R 4.0.0
 
-# semtree 0.9.13
+# semtree 0.9.13 (2018)
 
 - bugfix in invariance testing 
 - fix to remove warning from rpart plot (version 3.0.0 and above)
 
-# semtree 0.9.12
+# semtree 0.9.12 (2018)
 
 - bugfix in split evaluation procedure for cases where variables are named 'c', which crashed semtree
 - bugfix in varimpConvergencePlot: na.omit command logic was inverted
 
-# semtree 0.9.11
+# semtree 0.9.11 (2017)
 
 - semforests support for lavaan models including variable importance and partial dependence
 - crossvalidated likelihood supported with lavaan
 - plots of semtrees show correct #df when using focus parameters
 - added diversity functions
 
-# semtree 0.9.10
+# semtree 0.9.10 (2017)
 
 - added NEWS.md file
 - bugfix to partial dependence on variables of type factor

diff --git a/R/computePval_maxLR.R b/R/computePval_maxLR.R
@@ -1,18 +1,22 @@
 #' Wrapper function for computing the maxLR corrected p value
 #' from strucchange
 #' 
-#' @param maxLR: maximum of the LR test statistics
-#' @param q: number of free SEM parameters / degrees of freedom
-#' @param covariate: covariate under evaluation. This is important to get the level of
-#            measurement from the covariate and the bin size for ordinal and
-#            categorical covariates.
-#' @param from, to: numeric from interval (0, 1) specifying start and end of trimmed
-#           sample period. By default, to is 1 - from, i.e., with the default 
-#           from = 0.15 the first and last 15 percent of observations are
-#           trimmed. This only needed for continuous covariates.
-#' @param nrep: numeric. Number of replications used for simulating from the asymptotic 
-#       distribution (passed to efpFunctional). Only needed for ordinal
-#       covariates.
+#' @param maxLR maximum of the LR test statistics
+#' @param q number of free SEM parameters / degrees of freedom
+#' @param covariate covariate under evaluation. This is important to get the level of
+#'            measurement from the covariate and the bin size for ordinal and
+#'            categorical covariates.
+#' @param from numeric from interval (0, 1) specifying start of trimmed
+#'           sample period.  With the default 
+#'           from = 0.15 the first and last 15 percent of observations are
+#'           trimmed. This is only needed for continuous covariates.
+#' @param to numeric from interval (0, 1) specifying end of trimmed
+#'           sample period. By default, to is 1.
+#' @param nrep numeric. Number of replications used for simulating from the asymptotic 
+#'       distribution (passed to efpFunctional). Only needed for ordinal
+#'       covariates.
+#'
+#
 #' @author Manuel Arnold
 #' @return Numeric. p value for maximally selected LR statistic
 

diff --git a/R/fitSubmodels.R b/R/fitSubmodels.R
@@ -5,7 +5,8 @@
 #' @param subset1 Dataset for the first group model
 #' @param subset2 Dataset for the second group model
 #' @param control a \code{semtree.control} object
-#
+#' @param invariance fit models with invariant parameters if given. NULL otherwise (default).
+#' @param return.models boolean. Return the fitted models
 #' returns NA if fit fails
 #' 
 #' @export

diff --git a/R/semtree-package.R b/R/semtree-package.R
@@ -12,7 +12,7 @@
 #' @importFrom stats as.dist cmdscale coef cor cov logLik median pchisq qnorm runif var dist rnorm
 #' @importFrom graphics barplot legend lines pairs par plot strwidth text hist
 #' @import OpenMx
-#' @import bitops
+#' @importFrom bitops bitAnd
 #' @import rpart
 NULL
 

diff --git a/R/vcov_lavaan.R b/R/vcov_lavaan.R
@@ -1,8 +1,9 @@
 setMethod(f = "vcov", signature = signature(object = "lavaan"),
           definition = function(object, ...) {
+            N <- NA # this is to skip CRANs check error
             if (object@Model@eq.constraints) {
               K <- eval(parse(text = "lavaan:::lav_constraints_R2K(object@Model)"))
-              res <- solve(t(K) %*% lavInspect(object, what = "information.expected") %*% K * N) 
+              res <- solve(t(K) %*% lavaan::lavInspect(object, what = "information.expected") %*% K * N) 
             } else {
               res <- object@vcov$vcov
             }

diff --git a/README.Rmd b/README.Rmd
@@ -10,7 +10,7 @@ knitr::opts_chunk$set(
 
 [![DOI](http://zenodo.org/badge/76649538.svg)](https://zenodo.org/badge/latestdoi/76649538)
 [![cran version](http://www.r-pkg.org/badges/version/semtree)](https://cran.r-project.org/package=semtree)
-[![rstudio mirror downloads](http://cranlogs.r-pkg.org/badges/semtree)](https://github.com/metacran/cranlogs.app)
+[![rstudio mirror downloads](http://cranlogs.r-pkg.org/badges/semtree)](https://github.com/r-hub/cranlogs.app)
 [![Build Status](https://travis-ci.com/brandmaier/semtree.svg?branch=master)](https://travis-ci.com/brandmaier/semtree)
 [![License: GPL v3](https://img.shields.io/badge/License-GPLv3-blue.svg)](https://www.gnu.org/licenses/gpl-3.0)
 
@@ -39,7 +39,7 @@ devtools::install_github("brandmaier/semtree",force=TRUE, build_opts = c())
 
 Package documentation and use-cases with runnable R code can be found on our github pages: [https://brandmaier.github.io/semtree/](https://brandmaier.github.io/semtree/).
 
-You may also want to visit the semtree website: http://brandmaier.de/semtree
+You may also want to visit the semtree website: [https://brandmaier.de/semtree](https://brandmaier.de/semtree)
 
 Package vignettes (shipped with the package) contain documentation on how to use the package. Simply type this in R once you have loaded the package:
 

diff --git a/man/computePval_maxLR.Rd b/man/computePval_maxLR.Rd
diff --git a/man/fitSubmodels.Rd b/man/fitSubmodels.Rd
diff --git a/tests/vim.R b/tests/vim.R
@@ -68,7 +68,7 @@ lgcModel <- mxModel("Linear Growth Curve Model Path Specification",
                     )
 
 
-fr <- semforest(lgcModel, lgcm,control = semforest.control(num.trees = 10))
+fr <- semforest(lgcModel, lgcm,control = semforest.control(num.trees = 3))
 
 
 vimp <- varimp(fr)

diff --git a/vignettes/constraints.Rmd b/vignettes/constraints.Rmd
@@ -4,7 +4,7 @@ author: "Andreas M. Brandmaier"
 date: "`r Sys.Date()`"
 output: rmarkdown::html_vignette
 vignette: >
-  %\VignetteIndexEntry{constraints}
+  %\VignetteIndexEntry{Constraints in semtree}
   %\VignetteEngine{knitr::rmarkdown}
   %\VignetteEncoding{UTF-8}
 ---

diff --git a/vignettes/forests.Rmd b/vignettes/forests.Rmd
@@ -60,10 +60,10 @@ summary(result)
 
 ## Forest
 
-Create a forest control object that stores all tuning parameters of the forest
+Create a forest control object that stores all tuning parameters of the forest. Note that we use only 5 trees for demo purposes. Please increase the number in real applications.
 
 ```{r}
-control <- semforest.control(num.trees = 20)
+control <- semforest.control(num.trees = 5)
 print(control)
 ```
 

diff --git a/vignettes/getting-started.Rmd b/vignettes/getting-started.Rmd
@@ -2,7 +2,7 @@
 title: "Getting Started with the semtree package"
 output: rmarkdown::html_vignette
 vignette: >
-  %\VignetteIndexEntry{getting-started}
+  %\VignetteIndexEntry{Getting Started with the semtree package}
   %\VignetteEngine{knitr::rmarkdown}
   %\VignetteEncoding{UTF-8}
 ---

diff --git a/vignettes/score-based-tests.Rmd b/vignettes/score-based-tests.Rmd
@@ -4,7 +4,7 @@ author: "Andreas M. Brandmaier"
 date: "`r Sys.Date()`"
 output: rmarkdown::html_vignette
 vignette: >
-  %\VignetteIndexEntry{score-based-tests}
+  %\VignetteIndexEntry{Score-based Tests}
   %\VignetteEngine{knitr::rmarkdown}
   %\VignetteEncoding{UTF-8}
 ---
@@ -63,8 +63,15 @@ Use score-based tests to create the tree. Use Bonferroni-correction to adjust fo
 
 ```{r}
 library(semtree)
-ctrl = semtree.control(method="score", bonferroni = TRUE)
-tree = semtree(model=result, data=tree.data, control=ctrl)
+ctrl = semtree.control(
+  method="score", 
+  bonferroni = TRUE)
+```
+
+```{r}
+tree = semtree( model = result, 
+                data = tree.data, 
+                control=ctrl)
 ```
 
 Now let us plot the tree.

diff --git a/vignettes/semforest-focus.Rmd b/vignettes/semforest-focus.Rmd
@@ -4,7 +4,7 @@ author: "Andreas M. Brandmaier"
 date: "`r Sys.Date()`"
 output: rmarkdown::html_vignette
 vignette: >
-  %\VignetteIndexEntry{semtree-focus}
+  %\VignetteIndexEntry{Focus parameters in SEM forests}
   %\VignetteEngine{knitr::rmarkdown}
   %\VignetteEncoding{UTF-8}
 ---
@@ -87,7 +87,7 @@ Now, we are repeating the same analysis in a forest.
 
 forest <- semforest(model.biv, data=df.biv, 
                     constraints = list(focus.parameters=fp),
-                    control=semforest.control(num.trees=5))
+                    control=semforest.control(num.trees=5, control=semtree.control(method="score")))
 ```
 
 By default, we see that individual trees are fully grown (without a p-value threshold). The first split is according to `grp2` because it best explains the group differences. Subsequent splits are according to `grp1` even though the chi2 values are close to zero. They only appear because there is no p-value-based stopping criterion. 
@@ -96,16 +96,11 @@ By default, we see that individual trees are fully grown (without a p-value thre
 plot(forest$forest[[1]])
 ```
 
-Now, let us compare the new permutation scheme
+Now, let us investigate the permutation-based variable importance:
 
 ```{r}
 
 vim <- varimp(forest, method="permutationFocus")
 
-
-vim.old <- varimp(forest, method="permutation")
-
-
-plot(vim.old, main = "Standard")
-plot(vim, main="Focus correction")
+plot(vim, main="Variable Importance")
 ```