Version: 0.0.2

cite: Haghish, E. F. (in preparation). Integrating R machine learning algorithms in Stata using rcall 3.0: a tutorial for Stata users and developers

imputing missing data with Random Forest. this command utilizes the missForest package from R and embeds it in a Stata program using rcall package (Haghish, 2019). The missforest is a powerful command for EASILY imputing missing data.

more importantly, the command is also meant to be a tutorial for Stata developers, showing how to embed R into Stata and how to document Stata packages with Markdown language, using markdoc package package. visit the project on github and have a look at the source code!

Join the Resistance! Fork this repository on GitHub and contribute to its development or documentation.

The Syntax is under development!

missforest is a free and open-source R package for missing data imputation with Random Forest. The current Stata program embeds this package in a Stata program using rcall package (See Haghish, E. F. (2019) Seamless interactive language interfacing between R and Stata).

The missforest Stata program not only meant to provide the capabilities of Random Forest missing data imputation for Stata users, but also meant to be an example programs for developing Stata packages with R and rcall. In addition, the documentation of the package is written with markdoc package (See Haghish, E. F. (2019) Software documentation with markdoc 5.0) using the minimalistic Markdown language, which also can serve as a tutorial for the comunity of Stata developers how to save time on software documentation and make it friendlies and easier!

Here is an example of doing missing data imputation with the variablewise option, which provides the OOB error estimation for each imputed variable:

        . webuse mheart5 
        . missforest, variablewise
        . return list
        . matrix list r(oob)

E. F. Haghish
Department of Psychology
University of Oslo
[email protected]
machinelearning homepage
Package Updates on Twitter

This help file was dynamically produced by MarkDoc Literate Programming package