R6causal 0.8.3
- Changed the parameters of functions
counterfactual
and fairness
. Now the method to simulate counterfactuals must be always specified by the user and condition_type
is required if the method is “u_find”.
R6causal 0.8.2
- Fixed a bug in the interventions to missingness indicators (
rflist
).
R6causal 0.8.1
- The content of
simdata
and simdata_obs
revised when missing data mechanism is present.
- Interventions can be applied also to background variables (
uflist
) and missingness indicators (rflist
).
- Added counterfactual identification via R package
cfid
.
- Added tikz as an output format.
- Updated the method for creating the package documentation. (I do not really understand this but it works now.)
R6causal 0.8.0
- Added function
fairness
for fairness evaluation via counterfactual simulation.
- Added new default method “u_find” to function
counterfactual
to simulate counterfactuals for continuous and discrete variables.
- Added methods to R6 class
SCM
to add and remove vertices.
- Added methods to R6 class
SCM
to obtain parents, children, ancestors and descendants of a set of vertices.
- Added R6 class
LinearGaussianSCM
for random linear Gaussian SCMs.
- Added functions
analytic_linear_gaussian
, analytic_linear_gaussian_conditining
to work with linear Gaussian SCMs.
R6causal 0.7.0
- Added R6 class
ParallelWorld
to enable more general counterfactual simulations.
- Improved output control for simulations.
R6causal 0.6.1
- A minor change due to change in the igraph package.
R6causal 0.6.0
- The first version submitted to CRAN.
- Added a
NEWS.md
file to track changes to the package.