anthro: Anthropometric measures (weight and height)

In wbonat/mglm4twin: Multivariate Generalized linear Models for Twin Data

Anthropometric measures (weight and height)

Description

Anthropometric measures (weight and height) on 861 (327 DZ and 534 MZ) twin-pairs. Furthermore, we explore the flexibility of our proposed model class and model the dispersion. The data set is available as an example in the OpenMx package (Neale et al., 2016). We customize the data set for our needs, so make it available organized for use in the mlm4twin package.

• weight - Twin weight.

• height - Twin height.

• age - Twin age.

• Group - Twin zygosity (DZ - dizygotic; MZ - monozygotic).

• Twin - Twin code.

• Twin_pair - Code of twin within the pair (1 and 2).

Usage

data(anthro)

Format

a data.frame with 1722 records and 6 variables.

Source

Neale, M. C., Hunter, M. D., Pritikin, J. N., Zahery, M., Brick, T. R., Kirkpatrick, R. M., Estabrook, R., Bates, T. C., Maes, H. H. and Boker, S. M. (2016) OpenMx 2.0: Extended structural equation and statistical modeling. Psychometrika, 81, 535–549.

Bonat, W. H. and Hjelmborg, J. v. B. (2020) Multivariate Generalized Linear Models for Twin and Family data. to appear.

Examples

require(mglm4twin)
data(anthro, package="mglm4twin")
anthro$age <- (anthro$age - mean(anthro$age))/sd(anthro$age)
anthro$weight <- (anthro$weight - mean(anthro$weight))/sd(anthro$weight)
anthro$height <- (anthro$height - mean(anthro$height))/sd(anthro$height)
form_Wt <- weight ~ age + Group*Twin_pair
form_Ht <- height ~ age + Group*Twin_pair
biv0 <- list("formE1" = ~ age, "formE2" = ~ age, "formE12" = ~ age,
             "formA1" = ~ age, "formA2" = ~ age, "formA12" = ~ age,
             "formC1" = ~ age, "formC2" = ~ age, "formC12" = ~ age)
Z_biv0 <- mt_twin(N_DZ = 327, N_MZ = 534, n_resp = 2, model = "ACE",
                 formula = biv0, data = anthro)
control_initial <- list()
control_initial$regression <- list("R1" = c(0.13, 0.10, -0.20, -0.02, 0.037),
                                   "R2" = c(0.23, 0.01, -0.27, -0.11, 0.11))
control_initial$power <- list(c(0), c(0))
control_initial$tau <- c(0.15, 0, 0.12, rep(0,15))
fit_0 <- mglm4twin(linear_pred = c(form_Wt, form_Ht), matrix_pred = Z_biv0,
                   control_initial = control_initial,
                   control_algorithm = list(tuning = 0.5),
                   power_fixed = c(TRUE, TRUE), data = anthro)

wbonat/mglm4twin documentation built on Oct. 14, 2023, 9:37 p.m.