umx: umx_var – R documentation

Become an expert in R — Interactive courses, Cheat Sheets, certificates and more!

umx_var

Get variances from a df that might contain some non-numeric columns

Description

Pass in any dataframe and get variances despite some non-numeric columns. Cells involving these non-numeric columns are set to ordVar (default = 1).

Usage

umx_var(
  df,
  format = c("full", "diag", "lower"),
  use = c("complete.obs", "pairwise.complete.obs", "everything", "all.obs",
    "na.or.complete"),
  ordVar = 1,
  digits = NULL,
  strict = TRUE,
  allowCorForFactorCovs = FALSE
)

Arguments

`df`	A dataframe of raw data from which to get variances.
`format`	to return: options are c("full", "diag", "lower"). Defaults to full, but this is not implemented yet.
`use`	Passed to `cov()` - defaults to "complete.obs" (see param default for other options).
`ordVar`	The value to return at any ordinal columns (defaults to 1).
`digits`	digits to round output to (Ignored if NULL). Set for easy printing.
`strict`	Whether to allow non-ordered factors to be processed (default = FALSE (no)).
`allowCorForFactorCovs`	When ordinal data are present, use heterochoric correlations in affected cells, in place of covariances.

Value

mxModel()

References

https://tbates.github.io

Examples

tmp     = mtcars[,1:4]
tmp$cyl = ordered(mtcars$cyl) # ordered factor
tmp$hp  = ordered(mtcars$hp)  # binary factor
umx_var(tmp, format = "diag", ordVar = 1, use = "pair")
tmp2 = tmp[, c(1, 3)]
umx_var(tmp2, format = "diag")
umx_var(tmp2, format = "full")

data(myFADataRaw)
df = myFADataRaw[,c("z1", "z2", "z3")]
df$z1 = mxFactor(df$z1, levels = c(0, 1))
df$z2 = mxFactor(df$z2, levels = c(0, 1))
df$z3 = mxFactor(df$z3, levels = c(0, 1, 2))    
umx_var(df, format = "diag")
umx_var(df, format = "full", allowCorForFactorCovs=TRUE)

# Ordinal/continuous mix
data(twinData)
twinData= umx_scale_wide_twin_data(data=twinData,varsToScale="wt",sep= "")
# Cut BMI column to form ordinal obesity variables
obLevels   = c('normal', 'overweight', 'obese')
cuts       = quantile(twinData[, "bmi1"], probs = c(.5, .8), na.rm = TRUE)
twinData$obese1=cut(twinData$bmi1,breaks=c(-Inf,cuts,Inf),labels=obLevels)
twinData$obese2=cut(twinData$bmi2,breaks=c(-Inf,cuts,Inf),labels=obLevels)
# Make the ordinal variables into mxFactors
ordDVs = c("obese1", "obese2")
twinData[, ordDVs] = umxFactor(twinData[, ordDVs])
varStarts = umx_var(twinData[, c(ordDVs, "wt1", "wt2")], 
		format= "diag", ordVar = 1, use = "pairwise.complete.obs")

umx

Structural Equation Modeling and Twin Modeling in R

v4.10.10

GPL-3

Authors

Timothy C. Bates [aut, cre] (<https://orcid.org/0000-0002-1153-9007>), Gillespie Nathan [wit], Michael Zakharin [wit], Brenton Wiernik [ctb], Joshua N. Pritikin [ctb], Michael C. Neale [ctb], Hermine Maes [ctb]

Initial release

2021-11-30