ivmodel

ivmodel is a comprehensive R package for linear instrumental variables analysis with one endogenous variable (e.g. confounded treatment/policy/intervention). The package can handle large data, data from matched sets, and is designed to be robust to many data types (or mixtures of data types). The package contains the usual instrumental variables methods, methods for weak instruments, sensitivity analysis, and power calculations. The package also contains diagnostic tools to assess assumptions underlying IV analyses. All the methods in the software are outlined in Kang, Jiang, Zhao, and Small (2020) and Branson and Keele (2020).

To install this package in R from GitHub, run the following commands:

install.packages("devtools")
library(devtools) 
install_github("hyunseungkang/ivmodel")

To install this package in R from CRAN, run the following command:

install.packages("ivmodel")

Examples

library(ivmodel)

### Use Card (1995) data to estimate the effect of education on log earnings ###
# n: sample size; L: # of IVs; p: # of covariates 
# Y: n by 1 vector of outcomes (must be continuous)
# D: n by 1 vector of treatments (continuous or discrete)
# Z: n by L vector of instruments (continuous or discrete)
# X: n by L vector of instruments (continuous or discrete)
data(card.data)

# One Instrument Anaylsis with Proximity to 4yr College as IV#
Y=card.data[,"lwage"]
D=card.data[,"educ"] 
Z=card.data[,"nearc4"]
Xname=c("exper", "expersq", "black", "south", "smsa", "reg661",
"reg662", "reg663", "reg664", "reg665", "reg666", "reg667",
"reg668", "smsa66")
X=card.data[,Xname]

# Run command #
card.model1IV = ivmodel(Y=Y,D=D,Z=Z,X=X)
card.model1IV

# Obtain estimates of exogenous covariates' effects on outcome under #
# different k-class estimatos                                        #
coefOther(card.model1IV)

# Multiple IV Analysis with Proximito to 2yr and 4yr Colleges as IVs#
Z = card.data[,c("nearc4","nearc2")]
card.model2IV = ivmodel(Y=Y,D=D,Z=Z,X=X)
card.model2IV

# Use the formula environment
X = as.matrix(X)
card.modelFormula = ivmodelFormula(Y ~ D + X | Z + X)
card.modelFormula

References

Card, D. (1995). “Using Geographic Variations in College Proximity to Estimate the Return to Schooling.” In LN Christofides, EK Grant, R Swidinsky (eds.), Aspects of Labor Market Behaviour: Essays in Honour of John Vanderkamp. University of Toronto Press.

Kang, H., Jiang, Y., Zhao, Q., and Small, D. S. (2021). ivmodel: An R Package for Inference and Sensitivity Analysis of Instrumental Variables Models with One Endogenous Variable. Observational Studies. 7:1-24.

Branson, Z., Keele, J. (2020). Evaluating A Key Instrumental Variable Assumption Using Randomization Tests. American Journal of Epidemiology.