flps

library(flps)
#> Version: 1.0.0
#> 
#> It is a demo.
#> Acknowledgements. It is supported by the Institute of Education Sciences, U.S. Department of Education, through Grant R305D210036.

Fully Latent Principal Stratification (FLPS)¹

Fully Latent Principal Stratification (FLPS) is an extension of principal stratification.

Installation

Install the latest release from CRAN or git repo:

devtools::install_github("sooyongl/flps")
install.packages("flps")

Documentation is available here.
For compiling errors on Windows, see the relevant guide.

Example Usage

Data Generation and Preprocessing

Generate simulated rectangular data and convert it to a list of data suitable for the rstan package.
makeInpData() creates input data for running FLPS.
inp_data: a data frame containing all the data for FLPS. It is used in runFLPS function.

set.seed(10000)
inp_data <- flps::makeInpData(
  N       = 200,  # sample size
  R2Y     = 0.2,  # r^2 of outcome
  R2eta   = 0.5,  # r^2 of eta by one covariates
  omega   = 0.2,  # the effect of eta
  tau0    = 0.23, # direct effect
  tau1    = -0.16,# interaction effect between Z and eta
  betaL   = 0.2,
  betaY   = 0.4,
  lambda  = 0.8,  # the proportion of administered items
  nitem    = 10,   # the total number of items
  nfac    = 1,    # the number of latent factors
  lvmodel = 'rasch' # tag for latent variable model; case-sensitive (use lower-case letters)
)

# Input data matrix
data.table::data.table(inp_data)
#>                Y Z          X1        eta1 v1 v2 v3 v4 v5 v6 v7 v8 v9 v10
#>   1:  0.31384308 1 -0.74308228 -0.80447517  0  1  0  0  0  0  1  0  1   0
#>   2: -0.53385668 1  0.21197053 -0.29240295  1  0  1  1  1  1  0  0  0   1
#>   3: -0.01239108 1  0.34952377  0.39012726  1  1  1  0  0  1  1  1  0   1
#>   4:  0.05905941 1 -0.32419433  0.24014237  0  1  1  0  0  1  1  0  1   1
#>   5: -2.21654281 1 -1.80979412  0.52011169  0  1  0  0  0  1  1  1  1   0
#>  ---                                                                     
#> 196: -0.47559245 0  0.23763106 -0.27108910 NA NA NA NA NA NA NA NA NA  NA
#> 197:  0.46877629 0 -0.03646065  1.12609970 NA NA NA NA NA NA NA NA NA  NA
#> 198:  0.78717334 0  0.06867924  0.07008599 NA NA NA NA NA NA NA NA NA  NA
#> 199:  0.56380180 0  0.56467755  0.34826071 NA NA NA NA NA NA NA NA NA  NA
#> 200:  0.36809486 0  0.82158503 -0.35012492 NA NA NA NA NA NA NA NA NA  NA

Model Fitting with FLPS

The runFLPS function internally converts inp_data into the appropriate data format for rstan, utilizing the provided information before executing the FLPS analysis.

In order to optimize your workflow, consider utilizing the modelBuilder() function to pre-compile the Stan code, especially if you’re working with a relevant measurement model. This function compiles the Stan code and stores the resulting stanmodel object within the flps package directory.

Note: Should you encounter an error, ensure that your Rcpp package is updated to the latest version. Additionally, refreshing your R session is recommended after executing modelBuilder() to avoid potential conflicts. Upon subsequent usage of runFLPS(), this pre-compilation ensures the function operates more swiftly and efficiently by bypassing the compilation step, providing a smoother and faster analytical process. Once the Stan model is compiled, use importModel() to bring in the compiled Stan code. This code can then be provided to the compiled_stan argument in runFLPS. If this step is omitted, runFLPS() will compile the Stan code during each execution of FLPS.

modelBuilder(type = "rasch")
complied <- importModel(type = "rasch")

In case of errors, try the latest rstan and StanHeaders packages.

remove.packages(c("rstan", "StanHeaders"))
install.packages("rstan", repos = c("https://mc-stan.org/r-packages/", getOption("repos")))

Now, execute your FLPS model.

res <- runFLPS(
  inp_data = inp_data,
  # complied_stan = complied # if necessary
  outcome = "Y",
  trt = "Z",
  covariate = c("X1"),
  lv_type = "rasch",
  lv_model = "F =~ v1 + v2 + v3 + v4 + v5 + v6 + v7 + v8 + v9 + v10",
  stan_options = list(iter = 1000, warmup = 500, cores = 1, chains = 1)
)

Results

Retrieve summaries and visualize results with the following:

The flps_plot() shows the plot related to FLPS models


flps_plot(res, type = "causal")


flps_plot(res, type = "latent")