The hardware and bandwidth for this mirror is donated by METANET, the Webhosting and Full Service-Cloud Provider.
If you wish to report a bug, or if you are interested in having us mirror your free-software or open-source project, please feel free to contact us at mirror[@]metanet.ch.

raiseR 0.1.0

Initial release.

Collinearity diagnostics

• vif() – Variance Inflation Factor, with type = "O" (ordinary, correlation-matrix based) and type = "R" (robust, the RVIF of Jacob & Varadharajan, 2024). Accepts a formula, a fitted model object from this package (or an lm), or a numeric predictor matrix. rvif() is a convenience alias for vif(type = "R").
• cn() – Condition Number and full vector of Condition Indices, with ordinary (type = "O") and robust weighted (type = "R") versions.

Regression estimators

• raiseReg() – ordinary Raise Regression, with a sequential single-variable method (the default) and a simultaneous SVIF/QR method. Because raising leaves the column space of the design unchanged, its fitted values, residual standard error, R-squared and F-statistic are numerically identical to OLS, and its coefficient t-tests stay exactly valid.
• robRaise() – Robust Raise Regression with exact finite-sample inference (sandwich covariance, effective degrees of freedom, Satterthwaite-Welch correction), downweighting outliers via Stahel-Donoho projection outlyingness and Tukey’s biweight function.
• ridgeReg() / robRidge() – ordinary and robust Ridge Regression. robRidge() offers type = "MM" (IRLS with an MM seed) and type = "SDO" (Stahel-Donoho weighting, then ordinary ridge).
• liuReg() / robLiu() – ordinary and robust Liu Regression. The biasing parameter d defaults to the MSE-optimal dopt, with dmm and a manual value also available, and out-of-range estimates clipped to [0, 1] by default. robLiu() offers type = "MM" (following the MM-Liu estimator of Filzmoser & Kurnaz, with a closed-form biasing parameter) and type = "SDO".

Utilities and methods

• scaleDat() – scale a dataset by one of four conventions: classical (mean/sd), robust weighted (Stahel-Donoho/Tukey biweight), median/MADN, or min-max to [0, 1].
• print(), summary(), coef(), fitted(), residuals(), predict(newdata = ) and plot() methods for every fitted object, modelled on the corresponding lm methods.
• For raiseReg() (the exact, unbiased fit), the standard influence diagnostics hatvalues(), cooks.distance(), dfbetas() and covRatio(), plus lmtest::bptest() and car::ncvTest(), are supported and return values numerically identical to an equivalent lm() fit. robRaise() supports bptest() and ncvTest().

Dependencies

• The robust methods (robRaise(), robRidge(type = "SDO"), robLiu(type = "SDO"), vif(type = "R"), cn(type = "R"), scaleDat(type = "weighted")) require the mrfDepth package, which provides the projection outlyingness measure used in the published methodology.

These binaries (installable software) and packages are in development.
They may not be fully stable and should be used with caution. We make no claims about them.