cor_auto: Automatically compute an apppropriate correlation matrix

In qgraph: Graph Plotting Methods, Psychometric Data Visualization and Graphical Model Estimation

Automatically compute an apppropriate correlation matrix

Description

This is mainly a wrapper around Lavaan function lavCor (Rosseel, 2012) to compute a correlation matrix based on pychoric, polyserial and/or Pearson correlations. The wrapper removes all factors and searches for possible ordinal variabes. A variable is classified as ordinal if it is either ordered or if it consist of at most 7 unique integer values. After computing the correlations an additional check will be performed to see if the correlation matrix is positive definite.

Usage

cor_auto(data, select, detectOrdinal = TRUE, ordinalLevelMax = 7, npn.SKEPTIC = FALSE, 
          forcePD = FALSE, missing = "pairwise", verbose = TRUE)

Arguments

data	A data frame
select	Variables to select from the data frame (as used in subset)
detectOrdinal	Logical, should ordinal variables be detected? If FALSE only variables that are ordered are treated as ordinal variables
ordinalLevelMax	Integer specyfying the amount of unique integer values a variable should have to be classified as ordinal
npn.SKEPTIC	Logical, should the Nonparanormal SKEPTIC from the huge package be applied if the data is continuous? See huge.npn (Zhao, Liu, Roeder, Lafferty and Wasserman, 2014)
forcePD	If TRUE the function checks if the correlation matrix is positive definite. If the matrix is not positive definite nearPD from the Matrix package will be used (Bates and Maechler, 2014).
missing	Corresponds to the missing argument in lavCor
verbose	Logical, should information be printed to the console?

Value

A correlation matrix

Author(s)

Sacha Epskamp <[e-mail]>

References

Yves Rosseel (2012). lavaan: An R Package for Structural Equation Modeling. Journal of Statistical Software, 48(2), 1-36. URL http://www.jstatsoft.org/v48/i02/.

Tuo Zhao, Han Liu, Kathryn Roeder, John Lafferty and Larry Wasserman (2014). huge: High-dimensional Undirected Graph Estimation. R package version 1.2.6. http://CRAN.R-project.org/package=huge

Douglas Bates and Martin Maechler (2014). Matrix: Sparse and Dense Matrix Classes and Methods. R package version 1.1-3. http://CRAN.R-project.org/package=Matrix

Examples

## Not run: 
### Examples from lavCor (lavaan): ###

library("lavaan")

# Holzinger and Swineford (1939) example
HS9 <- HolzingerSwineford1939[,c("x1","x2","x3","x4","x5",
                                 "x6","x7","x8","x9")]

# Pearson correlations
cor_auto(HS9)

# ordinal version, with three categories
HS9ord <- as.data.frame( lapply(HS9, cut, 3, labels=FALSE) )

# polychoric correlations, two-stage estimation
cor_auto(HS9ord)

## End(Not run)

qgraph documentation built on Nov. 3, 2023, 5:07 p.m.