Software

I have written four R packages for the novel methodologies developed in my research. They are all available at CRAN.

PanelSelect

The PanelSelect package supports a series of panel sample selection models, where the first stage is a panel Probit model with individual random effects and the second stage can be a panel linear, Probit, Poisson, or Poisson log-normal model with individual random effects. A brief tutorial (vignette) for this package can be found here.

To install the latest release on CRAN:

install.packages("PanelSelect")

Endogeneity

The endogeneity package implements various recursive two-stage models to address the endogeneity issue in empirical research. These models can be used to address the endogeneity of treatment variables in observational studies or the endogeneity of mediators in mediation analysis. A brief tutorial (vignette) for this package can be found here.

To install the latest release on CRAN:

install.packages("endogeneity")

PanelCount

Panel count data are ubiquitous, such as the sales of products month by month and the views of videos day by day. The PanelCount package is a fast and scalable R package that supports random effects and sample selection models for panel count data. A brief tutorial (vignette) for this package can be found here.

To install the latest release on CRAN:

install.packages("PanelCount")

CoxPlus

Customers often make purchase decisions after exposure to multiple ads and it is important to identify the contribution of each ad in the path to conversion. The CoxPlus package implements a Cox model (proportional hazards model) that allows an event to have more than one causes. This model can be used to address the attribution problem in marketing campaigns.

To install the latest release on CRAN:

install.packages("CoxPlus")