Allow for easy-to-use testing or evaluating of linear equality and inequality restrictions about parameters and effects in (generalized) linear statistical models.
Restriktor is a free, open source R package for linear equality and inequality constrained statistical estimation, inference and evaluation for linear models.
Restriktor is implemented as an R package. This means that before installing restriktor, you should have installed a recent version (>= 4.0.0) of R. You can download the latest version of R from the R-project website.
R is a command line driven program. This means that it does not have a graphical user interface (GUI). Luckily, there are many good GUI's to make life easier, for example Rstudio, R Commander and RKWard.
Once you have installed R, the next step is to install restriktor. This can be done by typing in R:
install.packages("restriktor", dependencies = TRUE)
To check if the installation was successful, you can load the restriktor package and try for example:
library(restriktor)
# construct constraint syntax based on the factor level names
constraints <- 'GroupActive < GroupPassive < GroupControl < GroupNo'
Fit the unrestricted linear model, where "Age" is the response
variable and "Group" a factor with four treatment groups.
fit.ANOVA <- lm(Age ~ -1 + Group, data = ZelazoKolb1972)
# fit the restricted model
restr.ANOVA <- restriktor(fit.ANOVA, constraints = constraints)
# summary of the restricted parameter estimates
summary(restr.ANOVA)
# informative hypothesis tests
iht(restr.ANOVA)
# Generalized Order-Restricted Information Criterion (GORIC)
goric(restr.ANOVA, comparison = "complement")
If you can see the output, everything is set up and ready.
For more information see the restriktor website.