estimateLong: Estimate parameters.

In davidbolin/ngme: Linear Mixed Effects Models With Flexible Distributions

Estimate parameters.

Description

A function that estimates parameters by calling the "estimateLong_cpp()" function.

Usage

estimateLong(
  Y,
  locs = NULL,
  mixedEffect_list,
  measurment_list,
  processes_list,
  operator_list,
  step0 = 0.3,
  alpha = 0.3,
  learning_rate = 0,
  nBurnin_learningrate = NULL,
  nBurnin_base = 0,
  pSubsample = 1,
  polyak_rate = -1,
  subsample.type = 1,
  iter_start = 0,
  nPar_burnin = 0,
  pSubsample2 = 0.3,
  nIter = 10,
  nSim = 1,
  nBurnin = 10,
  silent = FALSE,
  seed = NULL,
  standardize.mixedEffects = FALSE,
  estimate_fisher = FALSE
)

Arguments

Y	A numeric list that contains outcome values.
locs	A numeric list that contains the timings at which the outcomes are collected.
mixedEffect_list	A list of inputs for random effects. noise The distribution of the mixed effects. B_random A list that contains the random effect covariates (needs to be matrix, can be NULL). B_fixed A list that contains the fixed effect covariates (needs to be matrix, can be NULL). beta_random Initial values for the parameters of the random effects (mean parameter) (if not specified set to zero). beta_fixed Initial values for the parameters of the fixed effects (if not specified set to zero). Sigma Initial values for the parameters of the variance-covariance matrix of the random effects (if not specified set to I ). nu Shape parameter for noise (NIG only) mu Shift parameter for noise (NIG only) U A list of inital values of the random effects. V A list of inital values of the variance effects.
measurment_list	A list of inputs for measurement error. sigma Measurement noise variance parameter. nu Shape parameter for noise (NIG only). Vs A list of inital values for the noise of the measurement.
processes_list	A list of inputs for the process. noise Distribution of the process. nu Shape parameter (for NIG or GAL). mu Asymmetry parameter (for NIG or GAL).
step0	A numeric value for stepsize for the optimizer; step0 / i^alpha.
alpha	A numeric value for stepsize for the optimizer; step0 / i^alpha.
learning_rate	A numeric value for the parameter of stochastic gradient.
nBurnin_learningrate	A numeric value until which the learning will not be started.
nBurnin_base	A numerical value for burn-in simulations that are performed for a subject that is sampled for the first time in the estimation method.
pSubsample	A numeric value for the portion of data to be used in each gradient iteration.
polyak_rate	A numeric value for moving average of parameters; -1: inactive, 0: pure mean.
subsample.type	A numeric value for the type of subsampling; 1: uniform sampling, 2: sample size weighted, 3: weighted sampling by gradient size, 4: grouped sub-sampler.
nPar_burnin	A numeric value; "M-step" updates will be used until this iteration.
pSubsample2	A numeric value for the portion of the data to be used in each gradient subsampling weighted by gradient.
nIter	A numeric value for the number of iteration that will be used by the stochastic gradient.
nSim	A numeric value for the number of samples of the Gibbs sampler to estimate the gradient.
silent	A logical value for printing the details of the iterations; "TRUE" indicates do not print, "FALSE" indicates print.
seed	A numerical value for starting the Gibbs samplers from fixed seed.
standardize.mixedEffects	A logical variable for standardising the covariates; "FALSE" indicates no standardisation, "TRUE" standardisation.
estimate_fisher	A logical variable for whether Fisher-Information matrix to be obtained; "FALSE" indicates do not obtain, "TRUE" obtain.

Details

This function calls "estimateLong_cpp()" internally. It is wrapped by "ngme", and is not advised to be used alone.

Value

A list of output.

See Also

ngme

Examples

  ## Not run: 
  data(srft_data)
  fit <- estimateLong(...)
  
## End(Not run)

davidbolin/ngme documentation built on Dec. 5, 2023, 11:48 p.m.