splines2 (version 0.2.1)

mSpline: M-Spline Basis for Polynomial Splines

Description

This function generates the monotone regression spline or simply called M-spline basis matrix for a polynomial spline.

Usage

mSpline(x, df = NULL, knots = NULL, degree = 3, intercept = FALSE, Boundary.knots = range(x), ...)

Arguments

x

The predictor variable. Missing values are allowed and will be returned as they were.

df

Degrees of freedom. One can specify df rather than knots, then the function chooses "df - degree" (minus one if there is an intercept) knots at suitable quantiles of x (which will ignore missing values). The default, NULL, corresponds to no inner knots, i.e., "degree - intercept".

knots

The internal breakpoints that define the spline. The default is NULL, which results in a basis for ordinary polynomial regression. Typical values are the mean or median for one knot, quantiles for more knots. See also Boundary.knots.

degree

Non-negative integer degree of the piecewise polynomial. The default value is 3 for cubic splines. Zero degree is allowed for piecewise constant basis.

intercept

If TRUE, an intercept is included in the basis; Default is FALSE.

Boundary.knots

Boundary points at which to anchor the M-spline basis. By default, they are the range of the non-NA data. If both knots and Boundary.knots are supplied, the basis parameters do not depend on x. Data can extend beyond Boundary.knots.

...

Optional arguments for future usage.

Value

A matrix of dimension length(x) by df = degree + length(knots) (plus one if intercept is included). Attributes that correspond to the arguments specified are returned for usage for predict.mSpline.

Details

It is an implementation of the close form M-spline basis based on relationship between M-spline basis and B-spline basis. In fact, M-spline basis is a rescaled version of B-spline basis. Internally, it calls function bSpline and generates a basis matrix for representing the family of piecewise polynomials with the specified interior knots and degree, evaluated at the values of x.

References

Ramsay, J. O. (1988). Monotone regression splines in action. Statistical science, 3(4), 425--441.

See Also

predict.mSpline for evaluation at given (new) values; bSpline for B-spline basis; iSpline for I-spline basis; cSpline for C-spline basis.

Examples

## Example given in the reference paper by Ramsay (1988)
library(graphics)
x <- seq(0, 1, by = .01)
knots <- c(0.3, 0.5, 0.6)
mMat <- mSpline(x, knots = knots, degree = 2, intercept = TRUE)
matplot(x, mMat, type = "l", ylab = "M-spline basis")
abline(v = knots, lty = 2, col = "gray")