OPSR 
Estimates ordinal probit switching regression (OPSR) models - a Heckman type selection model with an ordinal selection and continuous outcomes. Different model specifications are allowed for each treatment/regime.
Installation
Install from CRAN:
install.packages("OPSR")You can install the development version of OPSR from
GitHub with:
# install.packages("devtools")
devtools::install_github("dheimgartner/OPSR")Example
OPSR can be used whenever the ordinal treatment is not assigned
exogenously but self-selected and one is interested in the treatment
effect on a continuous outcome. The motivating example is telework
frequency (conceptually, the treatment) and vehicle miles driven (the
outcome of interest). We assume that two distinct processes lead people
to choose a certain telework frequency and how mobile they are. Further
and most importantly, the possibility of selection on unobservables
exists. I.e., unobserved factors (as part of the errors of the two
processes) might influence both the ordinal and continuous outcome. This
leads to error correlation which needs to be accounted for in the
modeling effort in spirit of Heckman.
library(OPSR)
#>
#> Please cite the 'OPSR' package as:
#> Heimgartner, D. and X. Wang (2024) OPSR: A package for estimating ordinal probit switching regression models in R. tbc.
#>
#> Wang, X. and P. L. Mokhtarian (2024) Examining the treatment effect of teleworking on vehicle-miles driven: Applying an ordered probit selection model and incorporating the role of travel stress, Transportation Research Part A, 186, 104072, doi:10.1016/j.tra.2024.104072.
#>
#> If you have questions, suggestions, or comments regarding the 'OPSR' package, please open an issue on https://github.com/dheimgartner/OPSR
#>
#> To see these entries in BibTeX format, use 'citation('OPSR')'
f <-
## ordinal and continuous outcome
twing_status | vmd_ln ~
## selection model
edu_2 + edu_3 + hhincome_2 + hhincome_3 +
flex_work + work_fulltime + twing_feasibility +
att_proactivemode + att_procarowning +
att_wif + att_proteamwork +
att_tw_effective_teamwork + att_tw_enthusiasm + att_tw_location_flex |
## outcome model NTW
female + age_mean + age_mean_sq +
race_black + race_other +
vehicle + suburban + smalltown + rural +
work_fulltime +
att_prolargehouse + att_procarowning +
region_waa |
## outcome model NUTW
edu_2 + edu_3 + suburban + smalltown + rural +
work_fulltime +
att_prolargehouse + att_proactivemode + att_procarowning |
## outcome model UTW
female + hhincome_2 + hhincome_3 +
child + suburban + smalltown + rural +
att_procarowning +
region_waa
fit <- opsr(f, telework_data, printLevel = 0)
texreg::screenreg(fit, beside = TRUE, include.pseudoR2 = TRUE, include.R2 = TRUE)
#>
#> ===============================================================================================
#> Structural Selection Outcome 1 Outcome 2 Outcome 3
#> -----------------------------------------------------------------------------------------------
#> kappa1 1.15 ***
#> (0.17)
#> kappa2 2.38 ***
#> (0.18)
#> sigma1 1.18 ***
#> (0.05)
#> sigma2 1.24 ***
#> (0.07)
#> sigma3 1.43 ***
#> (0.04)
#> rho1 0.07
#> (0.10)
#> rho2 0.13
#> (0.07)
#> rho3 0.30 ***
#> (0.07)
#> edu_2 0.24 0.22
#> (0.14) (0.34)
#> edu_3 0.40 ** 0.67 *
#> (0.13) (0.33)
#> hhincome_2 0.09 0.47
#> (0.11) (0.26)
#> hhincome_3 0.28 * 0.30
#> (0.11) (0.25)
#> flex_work 0.28 **
#> (0.10)
#> work_fulltime 0.25 * 0.44 *** 0.71 ***
#> (0.10) (0.13) (0.17)
#> twing_feasibility 0.13 ***
#> (0.01)
#> att_proactivemode 0.08 * -0.18 *
#> (0.04) (0.08)
#> att_procarowning -0.08 0.12 0.17 0.25 ***
#> (0.04) (0.07) (0.09) (0.06)
#> att_wif 0.12 **
#> (0.04)
#> att_proteamwork 0.09 *
#> (0.04)
#> att_tw_effective_teamwork 0.32 ***
#> (0.04)
#> att_tw_enthusiasm 0.09 *
#> (0.04)
#> att_tw_location_flex 0.08 *
#> (0.04)
#> (Intercept) 3.74 *** 2.42 *** 2.38 ***
#> (0.28) (0.39) (0.29)
#> female -0.21 * -0.37 ***
#> (0.11) (0.11)
#> age_mean 0.01 **
#> (0.00)
#> age_mean_sq -0.00
#> (0.00)
#> race_black -0.39
#> (0.24)
#> race_other -0.02
#> (0.18)
#> vehicle 0.13 **
#> (0.05)
#> suburban 0.01 0.45 * 0.29 *
#> (0.16) (0.17) (0.14)
#> smalltown 0.41 * 0.22 0.31
#> (0.18) (0.29) (0.28)
#> rural 0.49 * 0.82 * 0.85 **
#> (0.23) (0.32) (0.33)
#> att_prolargehouse 0.19 *** 0.16 *
#> (0.05) (0.08)
#> region_waa -0.24 * -0.27 *
#> (0.11) (0.11)
#> child 0.19 **
#> (0.06)
#> -----------------------------------------------------------------------------------------------
#> AIC 7189.51 7189.51 7189.51 7189.51 7189.51
#> BIC 7490.10 7490.10 7490.10 7490.10 7490.10
#> Log Likelihood -3538.76 -3538.76 -3538.76 -3538.76 -3538.76
#> Pseudo R^2 (EL) 0.49 0.49 0.49 0.49 0.49
#> Pseudo R^2 (MS) 0.46 0.46 0.46 0.46 0.46
#> R^2 0.24 0.24 0.24 0.24 0.24
#> Num. obs. 1584 1584 1584 1584 1584
#> ===============================================================================================
#> *** p < 0.001; ** p < 0.01; * p < 0.05