R Development Page
Contributed R Packages
Below is a list of all packages provided by project R Programming 2017 (Uni Innsbruck).
Important note for package
binaries: RForge provides these binaries only for
the most recent version of R, but not for older
versions. In order to successfully install the
packages provided on RForge, you have to switch
to the most recent version of R or, alternatively,
install from the package sources (.tar.gz).
GmmEst  Model Estimation by Generalized Method of Moments (GMM)


GMM estimation framework is provided. Onestep, twostep and iterative GMM estimation of a userspecified model is implemented. 

Version: 0.02 
Last change: 20170704 19:02:19+02 
Rev.: 205 

Download:
(.tar.gz) 
(.zip) 
Build status: Current 

R install command:
install.packages("GmmEst", repos="http://RForge.Rproject.org") 


bookMatrix  Model to build a booking matrix and analyze it


A prediction model for hotel bookings is provided.
The presented data is transformed in a cumulative booking matrix and then analyzed by a
linear regression model. 

Version: 0.01 
Last change: 20170620 18:06:56+02 
Rev.: 153 

Download:
(.tar.gz) 
(.zip) 
Build status: Current 

R install command:
install.packages("bookMatrix", repos="http://RForge.Rproject.org") 


ditree  Distributional Tree


Build a tree fitting a distributional model in each node.
As a framework the MOB algorithm is used to build the tree and a distribution is specified by handing over a gamlss.family object.
In each node of the tree a distributional model is fit via maximum likelihood and the resulting scores are used to find splits.
For each terminal node a set of distribution parameters is provided. 

Version: 0.01 
Last change: 20170804 16:13:23+02 
Rev.: 252 

Download:
(.tar.gz) 
(.zip) 
Build status: Failed to build 

R install command:
install.packages("ditree", repos="http://RForge.Rproject.org") 


enbin  Extended Negative Binomial (NB2) Regression


Negative binomial (NB2) regression models are provided.
These are applied to count data with a negative binomial distributed reponse.
The two distribution parameters (location and scale, i.e. theta)
may depend on covariates. A log link is employed for both the location
and the scale equation. 

Version: 0.02 
Last change: 20170625 12:51:50+02 
Rev.: 160 

Download:
(.tar.gz) 
(.zip) 
Build status: Current 

R install command:
install.packages("enbin", repos="http://RForge.Rproject.org") 


fgamma  Full Gamma Regression


Full gamma regression modeling the mu (mean) and sigma parameter as functions of explanatory variables using a "log" link for both parameters. 

Version: 0.10 
Last change: 20170724 14:50:03+02 
Rev.: 249 

Download:
(.tar.gz) 
(.zip) 
Build status: Current 

R install command:
install.packages("fgamma", repos="http://RForge.Rproject.org") 


hetprobit  Heteroscedastic Probit Regression Models


Heteroscedastic probit models are provided. In standard probit models the variance of the error term is assumed to be constant. Heteroscedastic probit models allow the error terms to vary systematically. 

Version: 0.10 
Last change: 20170720 11:13:37+02 
Rev.: 248 

Download:
(.tar.gz) 
(.zip) 
Build status: Current 

R install command:
install.packages("hetprobit", repos="http://RForge.Rproject.org") 


hetprobit2  Heteroscedastic Probit Regression Models v2


For nonlinear model like standard probit models, heteroscedasticity can have severe consequences: the maximum likelihood estimates of the parameters will be biased (in an unknown direction), as well as inconsistent (unless the likelihood function is modified to correctly take into account the precise form of heteroskedasticity). Greene points simply computing a robust covariance matrix for an otherwise inconsistent estimator does not give it redemption. Consequently, the virtue of a robust covariance matrix in this setting is unclear. Heteroscedastic probit models allow the error terms to vary systematically. 

Version: 0.01 
Last change: 20170529 16:59:57+02 
Rev.: 58 

Download:
(.tar.gz) 
(.zip) 
Build status: Current 

R install command:
install.packages("hetprobit2", repos="http://RForge.Rproject.org") 


htobit  Heteroscedastic Tobit Regression Models


Heteroscedastic tobit regression models are provided.
These are Gaussian regression models with a response variable leftcensored at zero
and both distribution parameters (the latent location and scale) parameters
can depend on covariates. An identity link is employed for the location
equation and a log link for the scale equation. 

Version: 0.10 
Last change: 20170714 01:22:22+02 
Rev.: 235 

Download:
(.tar.gz) 
(.zip) 
Build status: Current 

R install command:
install.packages("htobit", repos="http://RForge.Rproject.org") 


negbin1  Negative Binomial 1 Regression


Negative binomial 1 regression models are provided. The negative binomial distribution is an often used alternative to the Poisson model, especially when doubts arise regarding the independence of the underlying process and the inclusion of all relevant regressors. 

Version: 0.02 
Last change: 20170714 09:48:39+02 
Rev.: 239 

Download:
(.tar.gz) 
(.zip) 
Build status: Current 

R install command:
install.packages("negbin1", repos="http://RForge.Rproject.org") 


nndd  Nearest Neighbour Matching (NN) followed by a Linear Model with Difference in Differences (DD)


The treatment effect is estimated by applying nearest neighbours matching (NN) and difference in differences (DD). Nearest neighbours are matched by applying a GLM over an individual time span. In the following a liner model is estimated with a difference in differences setup. Each estimation (NN or DD) can depend on different covariates. Simple evaluation methods of the combined estimation are provided. 

Version: 0.02 
Last change: 20170706 20:38:45+02 
Rev.: 207 

Download:
(.tar.gz) 
(.zip) 
Build status: Current 

R install command:
install.packages("nndd", repos="http://RForge.Rproject.org") 


Build status codes
0  Current: the package is available for download. The corresponding package passed checks on the Linux and Windows platform without ERRORs.
1  Scheduled for build: the package has been recognized by the build system and provided in the staging area.
2  Building: the package has been sent to the build machines. It will be built and checked using the latest patched version of R. Note that it is included in a batch of several packages. Thus, this process will take some time to finish.
3  Failed to build: the package failed to build or did not pass the checks on the Linux and/or Windows platform. It is not made available since it does not meet the policies.
4  Conflicts: two or more packages of the same name exist. None of them will be built. Maintainers are asked to negotiate further actions.
5  Offline: the package is not available. The build system may be offline or the package maintainer did not trigger a rebuild (done e.g., via committing to the package repository).
If your package is not shown on this page or not building, then check the build system status report.