do_iterations: do lms iterations

Description

Do lms iterations

Usage

do_iterations(
  data.list,
  n = 10,
  max.it = 1000,
  method = "gamlss",
  prop.fam = 0.75,
  prop.subject = 1,
  age.min = 0,
  age.max = 18,
  age.int = 1/12,
  x2.min = 25,
  x2.max = 42,
  x2.int = 1/12,
  keep.models = F,
  dist = "BCCGo",
  mu.df = 4,
  sigma.df = 3,
  nu.df = 2,
  tau.df = 2,
  verbose = F,
  formula = NULL,
  sigma.formula = ~1,
  nu.formula = ~1,
  tau.formula = ~1,
  method.pb = "ML",
  trans.x = F,
  lim.trans = c(0, 1.5)
)

Value

list of lists for models and fitted parameters

Arguments

data.list: list of dataframes as returned by prepare_data
n: number of desired fits
max.it: maximum number of iterations that will be run
method: use vgam or gamlss
prop.fam: proportion of families to be sampled
prop.subject: proportion of subject to be sampled
age.min: lower bound of age
age.max: upper bound of age
age.int: stepwidth of the age variable
x2.min: minimum limit for the second predictor
x2.max: maximum limit for the second predictor
x2.int: interval length between knots saved
keep.models: indicator whether or not models in each iteration should be kept
dist: distribution used for the fitting process, has to be one of BCCGo, BCPEo, BCTo as they are accepted by lms()
mu.df: degree of freedem location parameter
sigma.df: degree of freedem spread parameter
nu.df: degree of freedem skewness parameter
tau.df: degree of freedem kurtosis parameter
verbose: whether or not information about sampling will be printed during while iterate
formula: formula for the location parameter
sigma.formula: formula for the sigma parameter
nu.formula: formula for the nu parameter
tau.formula: formula for the tau parameter
method.pb: GAIC or ML
trans.x: indicator wether age should be transformed or not
lim.trans: limits for the exponent of transformation of age

Author

Mandy Vogel

Details

function samples families, samples measurements (and subjects), fits the model for a given number of iterations