helper: Helper functions
In mboost: Model-Based Boosting

Description Usage Arguments Details See Also

These helper functions are used internally in mboost and are exported only to allow the package FDboost to use them.

## compute Ridge shrinkage parameter lambda from df or the other way round
df2lambda(X, df = 4, lambda = NULL, dmat = NULL, weights, XtX = NULL)

## hyper parameters for P-splines baselearner (including tensor product P-splines)
hyper_bbs(mf, vary, knots = 20, boundary.knots = NULL, degree = 3,
          differences = 2, df = 4, lambda = NULL, center = FALSE,
          cyclic = FALSE, constraint = "none", deriv = 0L)

## workhorse for fitting (ridge-penalized) baselearners
bl_lin(blg, Xfun, args)

`X`	the design matrix.
`df`	degrees of freedom. See `bbs`.
`lambda`	smoothing parameter. See `bbs`.
`dmat`	penalty matrix.
`weights`	regression weights.
`XtX`	(weighted) crossproduct of the design matrix.
`mf`	model frame.
`vary`	names of variables that specify varying coefficients. See argument `by` of `bbs`.
`knots, boundary.knots`	knots. See `bbs`.
`degree`	degree of the regression spline. See `bbs`.
`differences`	differences used in the penalty. See `bbs`.
`center`	use reparameterization? See `bbs`.
`cyclic`	use cyclic effects? See `bbs`.
`constraint`	type of constraint. See `bbs`.
`deriv`	See `bbs`.
`blg`	object of class `"blg"` that contains the model frame, etc.
`Xfun`	function to set up the model matrix given the arguments in `args`.
`args`	arguments. E.g. the result of a call to `hyper_bbs`.