lpme: R Packge for Dealing with Latent Predictor Measurement Error Under Identification Restrictions

Installation | Key Functions | References

lpme is an R package that provides tools for analyzing latent variable models with measurement error correction, using bootstrapping techniques for inference.

Package Installation

Within an R session, you can install the development version of lpme from GitHub with:

# install.packages("devtools") 
devtools::install_github("cjerzak/lpme/lpme-software")

Key Functions

lpme_OneRun

lpme_OneRun performs a single run of latent variable analysis with measurement error correction (no bootstrapping; 1 split sample partition):

# Generate data 
Yobs <- rnorm(1000)
ObservablesMat <- matrix(sample(c(0,1), 1000*10, replace = T), ncol = 10)

# One run of latent error correction method 
lpme::lpme_OneRun(Yobs, ObservablesMat, 
                  MakeObservablesGroupings = FALSE, seed = runif(1, 1, 10000))

lpme

lpme implements a bootstrap analysis for latent variable models with measurement error correction. We average over nPartition split sample partitions.

# Generate data 
Yobs <- rnorm(1000)
ObservablesMat <- matrix(sample(c(0,1), 1000*10, replace = T), ncol = 10)

# Latent error correction method, with partitioning and bootstrap 
results  <- lpme::lpme(Yobs, ObservablesMat, 
	     MakeObservablesGroupings = FALSE, nBoot = 32L, nPartition = 10L,
	     bootBasis = 1:length(Yobs), ReturnIntermediaries = TRUE,
	     seed = runif(1, 1, 10000)) 
		 
#View the corrected IV coefficient and its standard error
print(c(results$Corrected_IVRegCoef, results$Corrected_IVRegSE))

Contributing

Contributions to latenterror are welcome! Feel free to submit a pull request or open an issue.

Acknowledgements

We thank Guilherme Duarte, Jeff Lewis, Umberto Mignozzetti, Aaron Pancost, Erik Snowberg, Chris Tausanovitch, and participants of a panel at an MPSA panel for very helpful comments. We thank Major Valls for excellent research assistance.

References

Connor T. Jerzak, Stephen A. Jessee. Measurement Error in Latent Predictors: The Role of Identification Restrictions Working paper to be released summer 2024!