Roadmap

Roadmap

This page tracks functionality that exists in the R greybox package but is not yet ported (or only partially ported) to the Python package. It is the single source of truth for "is X available in Python?" — individual pages link here rather than scattering TBA markers. The authoritative status matrix is on Home; this page expands on the R-only entries.

Status: R-only = not in Python yet; partial = present in Python but missing features compared to R.

Modelling

Feature	R	Python	Status	Notes
Scale (variance) model	sm()	—	R-only	Dynamic model of the error scale.
Bootstrap of coefficients	coefbootstrap()	—	R-only
Three-parameter lognormal	dtplnorm() etc.	—	R-only	Extra distribution family.

Selection and evaluation

Feature	R	Python	Status	Notes
Regression Model Confidence Bounds test	rmcb()	—	R-only	Nemenyi/MCB-style comparison of methods.
Point information criteria	pAIC(), pAICc(), pBIC(), pBICc()	point_lik()	partial	Point-wise likelihood is in Python; the per-observation IC wrappers are not.

Association

Feature	R	Python	Status	Notes
Cramer's V	cramer()	—	R-only	Association for two categorical variables. See association.

Visualisation

Feature	R	Python	Status	Notes
Graph maker	graphmaker()	—	R-only	Actuals/fitted/forecast plot.
Spread plot	spread()	—	R-only	Scatterplot/boxplot matrix.
Table plot	tableplot()	—	R-only	Two-categorical-variable plot.

Utilities

Feature	R	Python	Status	Notes
DST detection	detectdst()	—	R-only
Leap-year detection	detectleap()	—	R-only
Polynomial products	polyprod()	—	R-only
DSR bootstrap	dsrboot()	—	R-only

Already ported (for reference)

Core functionality is available in both languages: ALM, stepwise, CALM, distributions, measures, rolling_origin, association (pcor/mcor/determination), diagnostics (outlier_dummy), manipulations (xreg_*, temporal_dummy), EDA (stick), Smoothers (lowess/supsmu), and AID (aid/aid_cat). See the parity matrix on Home.

References

See Resources for the full reference list.