OdatDT_10K: An example of a dataset in long format with random monitoring...

In osofr/estimtr: Streamlined Estimation for Static, Dynamic and Stochastic Treatment Regimes in Longitudinal Data

Description

Simulated dataset containing 10,000 i.i.d. observations organized in long format as person-time rows. The binary exposure is TI and binary outcome is Y.tplus1. See /tests/ for R code that generated this data as well as R code that uses stremr to analyze this data.

Usage

data(OdatDT_10K)

Format

A data frame with 10,000 observations and variables:

ID

Unique subject identifier

t

Interger for current time period, range 0-16

CVD

Baseline confounder (time invariant)

lastNat1

Time since last monitoring event, set to 0 when N[t-1]=0 and then added one for each new period where N[t] is 0.

highA1c

Time-varying confounder

TI

Binary exposure variable

C

Administrative censoring indicator, always set to 0 unless the end of study is reached (t==16)

N

The random indicator of being monitored (having a visit), simulated as a Bernoulli RV with P(N(t)=1)=0.5

Y.tplus1

Indicator of the survival event at t

gTI.dlow

Counterfactual exposure under static intervention - always treat

gTI.dhigh

Counterfactual exposure under dynamic intervention - treat only when highA1c is above 1 and the subject is being monitored

gPois3.yrly

Poisson probability of counterfactual monitoring indicator being equal to 1

gPois3.biyrly

Poisson probability of counterfactual monitoring indicator being equal to 1

gp05

Bernoulli probability of counterfactual monitoring indicator being equal to 1

osofr/estimtr documentation built on Jan. 25, 2022, 8:05 a.m.