ARZIMM: Auto-Regressive Zero-Inflated Mixed Model

In Hlch1992/Rprojects: Statistical modeling and inference of microbial interaction and stability for microbial dynamical systems

Description

This function allows you to fit the ARZIMM.

Usage

ARZIMM(
  phy,
  Varname,
  Conname = NA,
  Covname = NA,
  IDname = "ID",
  Tname = "Time",
  fdata = NULL,
  family = "Poisson",
  ...
)

Arguments

phy	A phyloseq-class experiment-level object example data. See phyExample
Varname	a vector of character string indicating the taxa names in the non-zero auto-regressive model
Conname	a vector of character string indicating the concomitant variable names in the zero state logit model
Covname	a vector of character string indicating the covariate names in the non-zero auto-regressive model
IDname	a character string indicating the subject ID. Default is ID
Tname	a character string indicating the time variable. Default is Time
fdata	a data frame containing all variables to be analysized
family	a character string indicating the distribtuion. default is Poisson
...	See ARZIMM.control

Value

an object of class "ARZIMMObject" is returned, which is a list with the ingredients of fit.

nwtable	the matrix of network table of fit
mseest	a list of mean square error: rmseest the root square of mean standard error rmseest the root square of mean pearson standardized error
paralist	a list of parameter estimates: beta the matrix of fixed effects for the non-zero auto-regressive model gamma the matrix of fixed effects for the zero state logit model sigma the measurement error standard deviation for both the non-zero auto-regressive model and the zero state logit model ciest the estimated random effects part of both the non-zero auto-regressive model and the zero state logit model
runtime	running time of the program
datalist	an object of class "ARZIMMData"
resultall	a list of parameter estimates of the fits with ingredients of the lambda
bootparapval	a list of p values obtain via bootstrap with componenets: betapval a vector of p values of fixed effects for the non-zero auto-regressive model gammapval a vector of p values of fixed effects for the zero state logit model sigmapval a vector of p values of the measurement error standard deviation for both the non-zero auto-regressive model and the zero state logit model
tunlist	the values of parameters used in the fits with components: lambdabeta the values of lambda used in the non-zero auto-regressive model lambdagamma the values of lambda used in the zero state logit model weight1all observation weights used in the non-zero auto-regressive model
parasetup	a list of parameters used to initial the ARZIMM program with components: initpara a list of initial parameter inputs; if the inputs are absent, default values are included lambda the values of lambda used in the fits. tunpara a list of tunning parameter inputs; if the inputs are absent, default values are included selectpara a list of selection parameter inputs; if the inputs are absent, default values are included

See Also

ARZIMM.control

Examples

data(phyExample)

Varname=colnames(otu_table(phyExample))
Conname=colnames(sample_data(phyExample))[1:6]
Tname=colnames(sample_data(phyExample))[7]
IDname=colnames(sample_data(phyExample))[8]

ARZIMMresult=ARZIMM::ARZIMM(phyExample,Varname = Varname,Conname = Conname,
IDname = IDname,Tname = Tname,bootpara=list(bootpval=TRUE,nboot=5))


data(sampledata)

Varname=colnames(sampledata)[1:20]
Conname=colnames(sampledata)[21:26]
Tname=colnames(sampledata)[27]
IDname=colnames(sampledata)[28]

ARZIMMresult=ARZIMM::ARZIMM(phy=NULL,Varname = Varname,Conname = Conname,fdata=sampledata,
IDname = IDname,Tname = Tname,bootpara=list(bootpval=TRUE,nboot=5))

Hlch1992/Rprojects documentation built on Feb. 10, 2020, 12:30 a.m.