umx (version 1.9.1)

umx_make_TwinData: Simulate twin data with control over A, C, and E parameters, as well as moderation of A.

Description

Makes MZ and DZ twin data, optionally with moderated A. By default, the three variance components must sum to 1.

See examples for how to use this: it is pretty flexible.

If you provide 2 varNames, they will be used. If you provide one, it will be expanded to var_T1 and var_T2

Note, if you want a power calculator, see here. You supply the number of pairs of each zygosity that wish to simulate (nMZpairs, nDZpairs), along with the values of AA, CC,and EE.

Shortcuts

You can omit nDZpairs. You can also give any 2 of A, C, or E and the function will add the value which makes the ACE total = 1.

Moderation

AA can take a list c(avg = .5, min = 0, max = 1). If specified will act like a moderated heritability, with average value = avg, and swinging down to min and up to max across 3 SDs of the moderator.

Usage

umx_make_TwinData(nMZpairs, nDZpairs = nMZpairs, AA = NULL, CC = NULL,
  EE = NULL, nThresh = NULL, sum2one = TRUE, varNames = "var",
  seed = NULL, empirical = FALSE, MZr = NULL, DZr = NULL)

Arguments

nMZpairs

Number of MZ pairs to simulate

nDZpairs

Number of DZ pairs to simulate (if omitted defaults to nMZpairs)

AA

value for A variance. Optionally a vector: c(avg= .5, min= 0, max= 1)

CC

value for C variance.

EE

value for E variance.

nThresh

If supplied, use as thresholds and return mxFactor output? (default is not too)

sum2one

Whether to enforce AA + CC + EE summing the one (default = TRUE)

varNames

name for var (defaults to 'var')

seed

Allows user to set.seed() if wanting reproducible dataset

empirical

Passed to mvrnorm

MZr

If MZr and DZr are set (default = NULL), the function simply returns dataframes of the request size and correlation

DZr

NULL

Value

• list of mzData and dzData dataframes containing T1 and T2 plus, if needed M1 and M2 (moderator values)

References

• https://github.com/tbates/umx, https://tbates.github.io

See Also

Other Data Functions: umxCovData, umxFactor, umxHetCor, umxPadAndPruneForDefVars, umx_as_numeric, umx_cont_2_quantiles, umx_cov2raw, umx_long2wide, umx_lower2full, umx_make_MR_data, umx_make_bin_cont_pair_data, umx_make_fake_data, umx_merge_CIs, umx_read_lower, umx_reorder, umx_residualize, umx_round, umx_scale_wide_twin_data, umx_scale, umx_swap_a_block, umx_wide2long, umx

Examples

# NOT RUN {
# =====================================================================
# = Basic Example, with all elements of std univariate data specified =
# =====================================================================
tmp = umx_make_TwinData(nMZpairs = 100, nDZpairs = 100, AA = .36, CC = .04, EE = .60)
# Show list of 2 data sets
str(tmp)
# = How to consume the built datasets =
mzData = tmp[[1]];
dzData = tmp[[2]];
cov(mzData); cov(dzData)
str(mzData); str(dzData); 

# Prefer to work in path coefficient values? (little a?)
tmp = umx_make_TwinData(200, AA = .6^2, CC = .2^2)
# Check the correlations
umxAPA(tmp[[1]]); umxAPA(tmp[[2]])

# =============
# = Shortcuts =
# =============

# Omit nDZpairs (equal numbers of both by default)
tmp = umx_make_TwinData(nMZpairs = 100, nDZpairs = 100, AA = .36, CC = .04, EE = .60)
tmp = umx_make_TwinData(100, AA = 0.5, CC = 0.3) # omit any one of A, C, or E (sums to 1)
cov(tmp[[1]])
# Not limited to unit variance
tmp = umx_make_TwinData(100, AA = 3, CC = 2, EE = 3, sum2one = FALSE) 
cov(tmp[[1]])

# =====================
# = Moderator Example =
# =====================

x = umx_make_TwinData(100, AA = c(avg = .7, min = 0, max = 1), CC = .55, EE = .63)
str(x)

# =====================
# = Threshold Example =
# =====================
tmp = umx_make_TwinData(100, AA = .6, CC = .2, nThresh = 3)
str(tmp)
umxAPA(tmp[[1]]); umxAPA(tmp[[2]])


# ========================
# = Just use MZr and DZr =
# ========================
tmp = umx_make_TwinData(100, MZr = .86, DZr= .60, varNames = "IQ")
umxAPA(tmp[[1]]); umxAPA(tmp[[2]])
# }