impsdata: Inpatient Multidimensional Psychiatric Scale (IMPS)

In ELCIC: The Empirical Likelihood-Based Consistent Information Criterion

Inpatient Multidimensional Psychiatric Scale (IMPS)

Description

The imps data frame has 1544 rows and 8 columns. The data is from National Institute of the Mental Health Schizophrenia Collaborative Study, where the effect of chlorpromazine, fluphenazine, or thioridazine treatment on the overall severity of the schizophrenia disorder is of interest.

Usage

data(impsdata)

Format

An object of class "list"

y

The binary outcomes indicating whether IMPS >=4, which is longitudinal dropout and missing at random. IMPS describes severity of the schizophrenia disorder (ranges from 0 to 7)

x

A full covariate matrix. It contains intercept, sex (1:male,0:female), drug (1: chlorphromazine, fluphenazine, or thioridazine treatment; 0: placebo), time: square root of the week covariate, and their two-way interactions.

x_mis

A covariate matrix for missing data model. It contains intercept, drug, time, and sex.

id

Patient ID

r

An indicator of the missingness (1: observed; 0: missing).

References

add here

Examples

data(impsdata)
id<-impsdata$id
r<-impsdata$r
data.real <- data.frame(id=id,y=impsdata$y, impsdata$x)
head(data.real,n=10)
# each participant has three records
time<-4
# the outcome is binary
family=binomial()
models <- list(y~Time, y~Drug,y~Time+Drug,
               y~Time*Drug,y~Time+Sex+Drug,
               y~Time+Sex+Drug+Time:Sex+Sex:Drug+Drug:Time)

model_mis<-r~Drug+Time+Sex

candidate.cor.sets<-c("exchangeable","independence","ar1")
#not run
#output_ELCIC<-ELCICwgee(models, candidate.cor.sets,data=data.real,model_mis,family,r,id,time)
#output_MLIC<-MLICwgee(models, candidate.cor.sets,data=data.real,model_mis,family,r,id,time)
#output_QICW<-QICWwgee(models, candidate.cor.sets,data=data.real,model_mis,family,r,id,time)

ELCIC documentation built on Feb. 16, 2023, 7:18 p.m.