lmestData: Data for 'LMest' functions

In LMest: Generalized Latent Markov Models

Data for LMest functions

Description

An object of class lmestData containing data in long format, some necessary information on the data structure and objects for the estimation functions.

Usage

lmestData(data, id = NULL, time = NULL,
          idAsFactor = TRUE, timeAsFactor = TRUE,
          responsesFormula = NULL, latentFormula = NULL,
          na.rm = FALSE, check.names = FALSE)

Arguments

data	a matrix or data frame in long format of observation
id	a numeric vector or a string indicating the column with the unit identifier. If NULL, the first column is considered
time	a numeric vector or a string indicating the column with the time occasions. If NULL, the second column is considered, and if the id is not NULL, the function will automatically add the column with the time occasions
idAsFactor	a logical value indicating whether or not the column with the ids is converted to a factor. (By default is set to TRUE)
timeAsFactor	a logical value indicating whether or not the column with the time occasions is converted in a factor. (By default is set to TRUE)
responsesFormula	A detailed description is given in lmest,lmestCont
latentFormula	A detailed description is given in lmest,lmestCont
na.rm	a logical value indicating whether or not the observation with at least a missing value is removed (By default is set to FALSE)
check.names	a logical value indicating whether or not the names of the variables are syntactically valid, and adjusted if necessary. (By default is set to FALSE)

Value

An object of class 'lmestData' with the following objects:

data	a data.frame object to use in the estimation functions
id	a integer vector with the unit identifier
time	a integer vector with the time occasions
n	the number of observation
TT	an integer value indicating number of time occasions
d	an interger value indicating the number of variables (columns except id and time)
Y	the response variables
Xmanifest	the variables affecting the measurement model if specified in responsesFormula
Xinitial	the variables affecting the initial probabilities of the latent model if specified in latentFormula
Xtrans	the variables affecting the transition probabilities of the latent model if specified in latentFormula

Author(s)

Francesco Bartolucci, Silvia Pandolfi, Fulvia Pennoni, Alessio Farcomeni, Alessio Serafini

Examples

data(data_long_cont)
str(data_long_cont)

## Data with continous resposes

dt <- lmestData(data = data_long_cont, id = "id",time="time")
str(dt)

## Summary of each variable and for each time

summary(dt)

## Summary of each variable

summary(dt, type = "cross")

## Summary of each variable by time

summary(dt, type = "year")

plot(dt)
plot(dt, typePlot = "sh")

#######################

## Not run: 

data("data_criminal_sim")

dt1 <- lmestData(data = data_criminal_sim, id = "id", time = "time")
str(dt1)

summary(dt1, varType = rep("d",ncol(dt1$Y)))

dt2 <- lmestData(data = data_criminal_sim, id = "id", time = "time",
                 responsesFormula = y1 + y2 ~ y3, latentFormula = ~ y7 + y8 | y9 + y10)
str(dt2)

## Summary for responses, covariates on the manifest distribution,
## covariates on intial and transition probabilities

summary(dt2, dataSummary = "responses",varType = rep("d",ncol(dt2$Y)))
summary(dt2, dataSummary = "manifest",varType = rep("d",ncol(dt2$Xmanifest)))
summary(dt2, dataSummary = "initial",varType = rep("d",ncol(dt2$Xinitial)))
summary(dt2, dataSummary = "transition",varType = rep("d",ncol(dt2$Xtrans)))


## End(Not run)

LMest documentation built on Aug. 27, 2023, 5:06 p.m.