stpm (version 1.1.2)

prepare_data: Data pre-processing for analysis with stochastic process model methodology.

Description

Data pre-processing for analysis with stochastic process model methodology.

Usage

prepare_data(x, y, col.id = NULL, col.status = NULL, col.age = NULL,
  col.age.event = NULL, covariates = NULL, interval = 1,
  verbose = FALSE)

Arguments

x

A path to the table with follow-up oservations (longitudinal study). Formats: csv, sas7bdat

y

A path to the table with vital statistics (mortality). File formats: csv, sas7bdat

col.id

A name of column containing subject ID. This ID should be the same in both longdat and vitstat tables. If not provided, the first column in the x and y will be used by default.

col.status

A name of the column containing status variable (0/1 which indicate alive/dead). If not provided - then the column #2 from the vital statistics dataset will be used.

col.age

A name of age column (also called 't1'). If not provided then the 3rd column from the longitudinal dataset (x) will be used.

col.age.event

A name of 'event' column. The event column indicates a time when the even occured (e.g. system failure). If not provided then the 3rd column from the vital statistics dataset will be used.

covariates

A list of covariates. If covariates not provided, then all columns from longitudinal table having index > 3 will be used as covariates.

interval

A number of breaks between observations for discrete model. Default = 1 unit of time.

verbose

A verbosing output indicator. Default=FALSE.

Value

• A list of two elements: first element contains a preprocessed data for continuous model, with arbitrary intervals between observations and second element contains a prepocessed data table for a discrete model (with constant intervals between observations).

Examples

library(stpm) 
data <- prepare_data(x=system.file("data","longdat.csv",package="stpm"), 
				   y=system.file("data","vitstat.csv",package="stpm"))
head(data[[1]])
head(data[[2]])