MCCM (version 0.1.0)

draw_correlation_matrix: Draw the Correlation Matrix

Description

Estimate the MCCM from dataframe and draw it with scatter plot of matrices (SPLOM). With bivariate scatter plots below the diagonal, histograms on the diagonal, and the polychoric correlation coefficients with standard errors above the diagonal. Correlation ellipses are drawn in the same graph. The red lines below the diagonal are the LOESS smoothed lines, fitting a smooth curve between two variables.

Usage

draw_correlation_matrix(
  data1,
  order_indx,
  pair_est = FALSE,
  MLE = FALSE,
  R0 = NULL,
  app = TRUE,
  korder = 2,
  max_iter = 1000,
  max_tol = 1e-08,
  show_log = FALSE
)

Value

the SPLOM plot.

Arguments

data1

a dataframe containing continuous or ordinal variable.

order_indx

a vector to indicate the ordinal variables.

pair_est

bool value, TRUE for pairwise estimation, FALSE for simultaneous estimation.

MLE

bool value, TRUE for maximum likelihood estimation, FALSE for IRLS (pairwise) or IGMM (simultaneous) estimation.

R0

the initial value for correlation vector, default Pearson correlation matrix.

app

bool value for approximation, TRUE for Legendre approximation, FALSE for common integral.

korder

the order of Legendre approximation.

max_iter

max iteration number for IGMM.

max_tol

max tolerance for iteration algorithm.

show_log

bool value, TRUE for showing calculation log.

See Also

MCCM_est, summary_MCCM_est

Examples

library(mvtnorm)
library(MASS)
library(polycor)
library(lavaan)
set.seed(1997)
n = 10000
rho12=0.3
rho13=0.4
rho14=0.5
rho23=0.6
rho24=0.7
rho34=0.8

R = matrix(c(1,rho12,rho13,rho14,rho12,1,rho23,rho24,rho13,rho23,1,rho34,
rho14,rho24,rho34,1),4,4)
indc = c(3,4)
thresholds = list(c(),c(),0,0)
data1 = gen_mixed(n=n,R=R,indc=indc,thresholds=thresholds)
data2 = data.frame(data1$observed)

# pairwise MLE estimation
draw_correlation_matrix(data2,indc,TRUE,TRUE)
# pairwise IRLS estimation
draw_correlation_matrix(data2,indc,TRUE,FALSE)
# simultaneous MLE estimation
draw_correlation_matrix(data2,indc,FALSE,TRUE)
# simultaneous IGMM estimation
draw_correlation_matrix(data2,indc,FALSE,FALSE)