Package Splines2': September 19, 2021

Package ‘splines2’
September 19, 2021

Title Regression Spline Functions and Classes
Version 0.4.5
Description Constructs basis matrix of B-splines, M-splines,
I-splines, convex splines (C-splines), periodic M-splines,
natural cubic splines, generalized Bernstein polynomials,
and their integrals (except C-splines) and derivatives
of given order by close-form recursive formulas.
It also contains a C++ head-only library integrated with Rcpp.
See Wang and Yan (2021) <doi:10.6339/21-JDS1020> for details.
Imports Rcpp, stats
LinkingTo Rcpp, RcppArmadillo
Suggests knitr, rmarkdown, tinytest
Depends R (>= 3.2.3)
VignetteBuilder knitr
License GPL (>= 3)
URL https://wwenjie.org/splines2,
https://github.com/wenjie2wang/splines2
BugReports https://github.com/wenjie2wang/splines2/issues
Encoding UTF-8
RoxygenNote 7.1.1
NeedsCompilation yes
Author Wenjie Wang [aut, cre] (<https://orcid.org/0000-0003-0363-3180>),
Jun Yan [aut] (<https://orcid.org/0000-0003-4401-7296>)
Maintainer Wenjie Wang <[email protected]>
Repository CRAN
Date/Publication 2021-09-19 13:20:02 UTC
1
2 bernsteinPoly
R topics documented:
bernsteinPoly . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2
bSpline . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4
cSpline . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6
dbs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8
deriv . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10
ibs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12
iSpline . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13
knots . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 15
mSpline . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 16
naturalSpline . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 20
predict . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 22
splines2 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 23
Index 25
bernsteinPoly Generalized Bernstein Polynomial Basis
Description
Returns a generalized Bernstein polynomial basis matrix of given degree over a specified range.
Usage
bernsteinPoly(
x,
degree = 3,
intercept = FALSE,
Boundary.knots = NULL,
derivs = 0L,
integral = FALSE,
...
)
Arguments
x The predictor variable taking values inside of the specified boundary. Missing
values are allowed and will be returned as they are.
degree A nonnegative integer representing the degree of the polynomials.
intercept If TRUE, the complete basis matrix will be returned. Otherwise, the first basis
will be excluded from the output.
Boundary.knots Boundary points at which to anchor the Bernstein polynomial basis. The default
value is NULL and the boundary knots is set internally to be range(x,na.rm =
TRUE).
bernsteinPoly 3
derivs A nonnegative integer specifying the order of derivatives. The default value is
0L for Bernstein polynomial basis functions.
integral A logical value. If TRUE, the integrals of the Bernstein polynomials will be
returned. The default value is FALSE.
... Optional arguments that are not used.
Value
A numeric matrix of dimension length(x) by degree + as.integer(intercept).
Examples
library(splines2)
x1 <- seq.int(0, 1, 0.01)

x2 <- seq.int(- 2, 2, 0.01)
## Bernstein polynomial basis matrix over [0, 1]

bMat1 <- bernsteinPoly(x1, degree = 4, intercept = TRUE)
## generalized Bernstein polynomials basis over [- 2, 2]

bMat2 <- bernsteinPoly(x2, degree = 4, intercept = TRUE)
op <- par(mfrow = c(1, 2), mar = c(2.5, 2.5, 0.2, 0.1), mgp = c(1.5, 0.5, 0))
matplot(x1, bMat1, type = "l", ylab = "y")
matplot(x2, bMat2, type = "l", ylab = "y")
## the first and second derivative matrix

d1Mat1 <- bernsteinPoly(x1, degree = 4, derivs = 1, intercept = TRUE)
par(mfrow = c(2, 2))

matplot(x1, d1Mat1, type = "l", ylab = "y")
## reset to previous plotting settings

par(op)
## or use the deriv method

all.equal(d1Mat1, deriv(bMat1))
all.equal(d2Mat1, deriv(bMat1, 2))
## the integrals
iMat1 <- bernsteinPoly(x1, degree = 4, integral = TRUE, intercept = TRUE)
iMat2 <- bernsteinPoly(x2, degree = 4, integral = TRUE, intercept = TRUE)
all.equal(deriv(iMat1), bMat1, check.attributes = FALSE)
all.equal(deriv(iMat2), bMat2, check.attributes = FALSE)
4 bSpline
bSpline B-Spline Basis for Polynomial Splines
Description
Generates the B-spline basis matrix representing the family of piecewise polynomials with the spec-
ified interior knots, degree, and boundary knots, evaluated at the values of x.
Usage
bSpline(
x,
df = NULL,
knots = NULL,
degree = 3L,
intercept = FALSE,
derivs = 0L,
integral = FALSE,
...
)
Arguments
x The predictor variable. Missing values are allowed and will be returned as they
are.
df Degree of freedom that equals to the column number of the returned matrix.
One can specify df rather than knots, then the function chooses df -degree
-as.integer(intercept) internal knots at suitable quantiles of x ignoring
missing values and those x outside of the boundary. If internal knots are speci-
fied via knots, the specified df will be ignored.
knots The internal breakpoints that define the splines. 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.
degree A nonnegative integer specifying the degree of the piecewise polynomial. The
default value is 3 for cubic splines. Zero degree is allowed for piecewise constant
basis functions.
Boundary.knots Boundary points at which to anchor the splines. By default, they are the range
of x excluding NA. If both knots and Boundary.knots are supplied, the basis
parameters do not depend on x. Data can extend beyond Boundary.knots.
derivs A nonnegative integer specifying the order of derivatives of B-splines. The de-
fault value is 0L for B-spline basis functions.
bSpline 5
integral A logical value. If TRUE, the corresponding integrals of spline basis functions
will be returned. The default value is FALSE.
Details
This function extends the bs() function in the splines package for B-spline basis by allowing
piecewise constant (left-closed and right-open except on the right boundary) spline basis of degree
zero.
Value
A numeric matrix of length(x) rows and df columns if df is specified or length(knots) + degree
+ as.integer(intercept) columns if knots are specified instead. Attributes that correspond to
the arguments specified are returned mainly for other functions in this package.
References
De Boor, Carl. (1978). A practical guide to splines. Vol. 27. New York: Springer-Verlag.
See Also
dbs for derivatives of B-splines; ibs for integrals of B-splines;
Examples
library(splines2)
x <- seq.int(0, 1, 0.01)

knots <- c(0.3, 0.5, 0.6)
## cubic B-splines
bsMat <- bSpline(x, knots = knots, degree = 3, intercept = TRUE)
op <- par(mar = c(2.5, 2.5, 0.2, 0.1), mgp = c(1.5, 0.5, 0))
matplot(x, bsMat, type = "l", ylab = "Cubic B-splines")
abline(v = knots, lty = 2, col = "gray")

par(op)
## the first derivaitves

d1Mat <- deriv(bsMat)
## the second derivaitves

d2Mat <- deriv(bsMat, 2)
## evaluate at new values

predict(bsMat, c(0.125, 0.801))
6 cSpline
cSpline C-Spline Basis for Polynomial Splines
Description
Generates the convex regression spline (called C-spline) basis matrix by integrating I-spline basis
for a polynomial spline or the corresponding derivatives.
Usage
cSpline(
x,
df = NULL,
knots = NULL,
degree = 3L,
intercept = TRUE,
derivs = 0L,
scale = TRUE,
...
)
Arguments
are.
degree The degree of C-spline defined to be the degree of the associated M-spline in-
stead of actual polynomial degree. For example, C-spline basis of degree 2 is
defined as the scaled double integral of associated M-spline basis of degree 2.
intercept If TRUE by default, all of the spline basis functions are returned. Notice that
when using C-Spline for shape-restricted regression, intercept = TRUE should
be set even when an intercept term is considered additional to the spline basis in
the model.
derivs A nonnegative integer specifying the order of derivatives of C-splines. The de-
fault value is 0L for C-spline basis functions.
cSpline 7
scale A logical value indicating if scaling C-splines is required. If TRUE by default,

each C-spline basis is scaled to have unit height at right boundary knot. The
corresponding I-spline and M-spline produced by deriv methods will be scaled
to the same extent.
Details
It is an implementation of the closed-form C-spline basis derived from the recursion formula of
I-splines and M-splines.
Value
References
Meyer, M. C. (2008). Inference using shape-restricted regression splines. The Annals of Applied
Statistics, 2(3), 1013–1033.
See Also
iSpline for I-splines; mSpline for M-splines.
Examples
library(splines2)
x <- seq.int(0, 1, 0.01)

knots <- c(0.3, 0.5, 0.6)
### when 'scale = TRUE' (by default)

csMat <- cSpline(x, knots = knots, degree = 2)
op <- par(mar = c(2.5, 2.5, 0.2, 0.1), mgp = c(1.5, 0.5, 0))
matplot(x, csMat, type = "l", ylab = "C-spline basis")
isMat <- deriv(csMat)
msMat <- deriv(csMat, derivs = 2)
matplot(x, isMat, type = "l", ylab = "scaled I-spline basis")
matplot(x, msMat, type = "l", ylab = "scaled M-spline basis")

par(op)
### when 'scale = FALSE'

csMat <- cSpline(x, knots = knots, degree = 2, scale = FALSE)
## the corresponding I-splines and M-splines (with same arguments)

8 dbs
isMat <- iSpline(x, knots = knots, degree = 2)

msMat <- mSpline(x, knots = knots, degree = 2, intercept = TRUE)
## or using deriv methods (more efficient)

isMat1 <- deriv(csMat)
msMat1 <- deriv(csMat, derivs = 2)
## equivalent
stopifnot(all.equal(isMat, isMat1, check.attributes = FALSE))
stopifnot(all.equal(msMat, msMat1, check.attributes = FALSE))
dbs Derivatives of B-Splines
Description
Produces the derivatives of given order of B-splines.
Usage
dbs(
x,
derivs = 1L,
df = NULL,
knots = NULL,
degree = 3L,
intercept = FALSE,
...
)
Arguments
are.
derivs A positive integer specifying the order of derivative. The default value is 1L for
the first derivative.
dbs 9
basis functions.
Details
This function provides a more user-friendly interface and a more consistent handling for NA’s than
splines::splineDesign() for derivatives of B-splines. The implementation is based on the
closed-form recursion formula. At knots, the derivative is defined to be the right derivative except
at the right boundary knot.
Value

References
See Also
bSpline for B-splines; ibs for integrals of B-splines.
Examples
library(splines2)
x <- seq.int(0, 1, 0.01)
knots <- c(0.2, 0.4, 0.7)
## the second derivative of cubic B-splines with three internal knots
dMat <- dbs(x, derivs = 2L, knots = knots, intercept = TRUE)
## compare with the results from splineDesign

ord <- attr(dMat, "degree") + 1L
bKnots <- attr(dMat, "Boundary.knots")
aKnots <- c(rep(bKnots[1L], ord), knots, rep(bKnots[2L], ord))
res <- splines::splineDesign(aKnots, x = x, derivs = 2L)
stopifnot(all.equal(res, dMat, check.attributes = FALSE))
10 deriv
deriv Derivatives of Spline Basis Functions
Description
Returns derivatives of given order for the given spline basis functions.
Usage
## S3 method for class 'bSpline2'
deriv(expr, derivs = 1L, ...)
## S3 method for class 'dbs'

## S3 method for class 'ibs'

## S3 method for class 'mSpline'

## S3 method for class 'iSpline'

## S3 method for class 'cSpline'

## S3 method for class 'bernsteinPoly'

## S3 method for class 'naturalSpline'

Arguments
expr Objects of class bSpline2, ibs, mSpline, iSpline, cSpline, bernsteinPoly
or naturalSpline with attributes describing knots, degree, etc.
derivs A positive integer specifying the order of derivatives. By default, it is 1L for the
first derivatives.
Details
At knots, the derivative is defined to be the right derivative except at the right boundary knot. By
default, the function returns the first derivatives. For derivatives of order greater than one, nested
function calls such as deriv(deriv(expr)) are supported but not recommended. For a better
performance, argument derivs should be specified instead.
deriv 11
This function is designed for objects produced by this package. It internally extracts necessary
specification about the spline/polynomial basis matrix from its attributes. Therefore, the function
will not work if the key attributes are not available after some operations.
Value
A numeric matrix of the same dimension with the input expr.
Examples
library(splines2)
x <- c(seq.int(0, 1, 0.1), NA) # NA's will be kept.

knots <- c(0.3, 0.5, 0.6)
## helper function
stopifnot_equivalent <- function(...) {
stopifnot(all.equal(..., check.attributes = FALSE))
}
## integal of B-splines and the corresponding B-splines integrated

ibsMat <- ibs(x, knots = knots)
bsMat <- bSpline(x, knots = knots)
## the first derivative

d1Mat <- deriv(ibsMat)
stopifnot_equivalent(bsMat, d1Mat)
## the second derivative

d2Mat1 <- deriv(bsMat)
d2Mat2 <- deriv(ibsMat, derivs = 2L)
stopifnot_equivalent(d2Mat1, d2Mat2)
## nested calls are supported

d2Mat3 <- deriv(deriv(ibsMat))
stopifnot_equivalent(d2Mat2, d2Mat3)
## C-splines, I-splines, M-splines and the derivatives

csMat <- cSpline(x, knots = knots, intercept = TRUE, scale = FALSE)
isMat <- iSpline(x, knots = knots, intercept = TRUE)
stopifnot_equivalent(isMat, deriv(csMat))
msMat <- mSpline(x, knots = knots, intercept = TRUE)

stopifnot_equivalent(msMat, deriv(isMat))
stopifnot_equivalent(msMat, deriv(csMat, 2))
stopifnot_equivalent(msMat, deriv(deriv(csMat)))
dmsMat <- mSpline(x, knots = knots, intercept = TRUE, derivs = 1)

stopifnot_equivalent(dmsMat, deriv(msMat))
stopifnot_equivalent(dmsMat, deriv(isMat, 2))
stopifnot_equivalent(dmsMat, deriv(deriv(isMat)))
stopifnot_equivalent(dmsMat, deriv(csMat, 3))
stopifnot_equivalent(dmsMat, deriv(deriv(deriv(csMat))))
12 ibs
ibs Integrals of B-Splines
Description
Generates basis matrix for integrals of B-splines.
Usage
ibs(
x,
df = NULL,
knots = NULL,
degree = 3,
intercept = FALSE,
...
)
Arguments
are.
basis functions.
Details
The implementation is based on the closed-form recursion formula.
iSpline 13
Value

References
See Also
bSpline for B-splines; dbs for derivatives of B-splines;
Examples
library(splines2)
x <- seq.int(0, 1, 0.01)

knots <- c(0.2, 0.4, 0.7, 0.9)
ibsMat <- ibs(x, knots = knots, degree = 1, intercept = TRUE)
## get the corresponding B-splines by bSpline()

bsMat0 <- bSpline(x, knots = knots, degree = 1, intercept = TRUE)
## or by the deriv() method
bsMat <- deriv(ibsMat)
stopifnot(all.equal(bsMat0, bsMat, check.attributes = FALSE))
## plot B-spline basis with their corresponding integrals

op <- par(mfrow = c(1, 2))
matplot(x, bsMat, type = "l", ylab = "B-spline basis")
matplot(x, ibsMat, type = "l", ylab = "Integral of B-spline basis")

par(op)
iSpline I-Spline Basis for Polynomial Splines
Description
Generates the I-spline (integral of M-spline) basis matrix for a polynomial spline or the correspond-
ing derivatives of given order.
14 iSpline
Usage
iSpline(
x,
df = NULL,
knots = NULL,
degree = 3L,
intercept = TRUE,
derivs = 0L,
...
)
Arguments
are.
degree The degree of I-spline defined to be the degree of the associated M-spline instead
of actual polynomial degree. For example, I-spline basis of degree 2 is defined
as the integral of associated M-spline basis of degree 2.
intercept If TRUE by default, all of the spline basis functions are returned. Notice that when
using I-Spline for monotonic regression, intercept = TRUE should be set even
when an intercept term is considered additional to the spline basis functions.
derivs A nonnegative integer specifying the order of derivatives of I-splines.
Details
It is an implementation of the closed-form I-spline basis based on the recursion formula given by
Ramsay (1988).
Value
knots 15
References
Ramsay, J. O. (1988). Monotone regression splines in action. Statistical Science, 3(4), 425–441.
See Also
mSpline for M-splines; cSpline for C-splines;
Examples
library(splines2)
## Example given in the reference paper by Ramsay (1988)

x <- seq.int(0, 1, by = 0.01)
knots <- c(0.3, 0.5, 0.6)
op <- par(mar = c(2.5, 2.5, 0.2, 0.1), mgp = c(1.5, 0.5, 0))
matplot(x, isMat, type = "l", ylab = "I-spline basis")

par(op)
## the derivative of I-splines is M-spline

msMat1 <- iSpline(x, knots = knots, degree = 2, derivs = 1)
msMat2 <- mSpline(x, knots = knots, degree = 2, intercept = TRUE)
stopifnot(all.equal(msMat1, msMat2))
knots Extract Knots from the Given Object
Description
Methods for the generic function knots from the stats package to obtain internal or boundary knots
from the objects produced by this package.
Usage
knots(Fn, type = c("internal", "boundary"), ...)



16 mSpline




Arguments
Fn An splines2 object produced by this package.
type A character vector of length one indicating the type of knots to return. The avail-
able choices are "internal" for internal knots and "Boundary" for boundary
knots.
... Optional arguments that are not used now.
Value
A numerical vector.
Examples
library(splines2)
set.seed(123)
x <- rnorm(100)
## B-spline basis
bsMat <- bSpline(x, df = 8, degree = 3)
## extract internal knots placed based on the quantile of x

(internal_knots <- knots(bsMat))
## extract boundary knots placed based on the range of x

boundary_knots <- knots(bsMat, type = "boundary")
all.equal(boundary_knots, range(x))
mSpline M-Spline Basis for Polynomial Splines

mSpline 17
Description
Generates the basis matrix of regular M-spline, periodic M-spline, and the corresponding integrals
and derivatives.
Usage
mSpline(
x,
df = NULL,
knots = NULL,
degree = 3L,
intercept = FALSE,
periodic = FALSE,
derivs = 0L,
integral = FALSE,
...
)
Arguments
are.
One can specify df rather than knots. For M-splines, the function chooses
df -degree -as.integer(intercept) internal knots at suitable quantiles of x
ignoring missing values and those x outside of the boundary. For periodic M-
spline (periodic = TRUE), df -as.integer(intercept) internal knots will be
chosen at suitable quantiles of x relative to the beginning of the cyclic intervals
they belong to (see Examples) and the number of internal knots must be greater
or equal to the specified degree -1. If internal knots are specified via knots,
the specified df will be ignored.
mean or median for one knot, quantiles for more knots. For periodic splines
(periodic = TRUE), the number of knots must be greater or equal to the specified
degree -1.
basis functions.
of x excluding NA. If both knots and Boundary.knots are supplied, the ba-
sis parameters do not depend on x. Data can extend beyond Boundary.knots.
For periodic splines (periodic = TRUE), the specified boundary knots define the
cyclic interval.
18 mSpline
periodic A logical value. If TRUE, the periodic splines will be returned instead of regular
M-splines. The default value is FALSE.
derivs A nonnegative integer specifying the order of derivatives of M-splines. The
default value is 0L for M-spline basis functions.
integral A logical value. If TRUE, the corresponding integrals of spline basis functions
will be returned. The default value is FALSE. For periodic splines, the integral of
each basis is integrated from the left boundary knot.
Details
This function contains an implementation of the closed-form M-spline basis based on the recursion
formula given by Ramsay (1988) or periodic M-spline basis following the procedure producing
periodic B-splines given in Piegl and Tiller (1997). For monotone regression, one can use I-splines
(see iSpline) instead of M-splines.
Value
A numeric matrix of length(x) rows and df columns if df is specified. If knots are specified
instead, the output matrix will consist of length(knots) + degree + as.integer(intercept)
columns if periodic = FALSE, or length(knots) + as.integer(intercept) columns if periodic
= TRUE. Attributes that correspond to the arguments specified are returned for usage of other func-
tions in this package.
References
Ramsay, J. O. (1988). Monotone regression splines in action. Statistical science, 3(4), 425–441.
Piegl, L., & Tiller, W. (1997). The NURBS book. Springer Science \& Business Media.
See Also
bSpline for B-splines; iSpline for I-splines; cSpline for C-splines.
Examples
library(splines2)
### example given in the reference paper by Ramsay (1988)

x <- seq.int(0, 1, 0.01)
knots <- c(0.3, 0.5, 0.6)
msMat <- mSpline(x, knots = knots, degree = 2, intercept = TRUE)
op <- par(mar = c(2.5, 2.5, 0.2, 0.1), mgp = c(1.5, 0.5, 0))
matplot(x, msMat, type = "l", ylab = "y")
## derivatives of M-splines
dmsMat <- mSpline(x, knots = knots, degree = 2,
intercept = TRUE, derivs = 1)
mSpline 19
## or using the deriv method

dmsMat1 <- deriv(msMat)
stopifnot(all.equal(dmsMat, dmsMat1, check.attributes = FALSE))
### periodic M-splines

x <- seq.int(0, 3, 0.01)
bknots <- c(0, 1)
pMat <- mSpline(x, knots = knots, degree = 3, intercept = TRUE,
Boundary.knots = bknots, periodic = TRUE)
## integrals
iMat <- mSpline(x, knots = knots, degree = 3, intercept = TRUE,
Boundary.knots = bknots, periodic = TRUE, integral = TRUE)
## first derivatives by "derivs = 1"
dMat1 <- mSpline(x, knots = knots, degree = 3, intercept = TRUE,
Boundary.knots = bknots, periodic = TRUE, derivs = 1)
## first derivatives by using the deriv() method
dMat2 <- deriv(pMat)
par(mfrow = c(2, 2))

matplot(x, pMat, type = "l", ylab = "Periodic Basis")
abline(v = seq.int(0, max(x)), lty = 2, col = "grey")
matplot(x, iMat, type = "l", ylab = "Integrals from 0")
abline(v = seq.int(0, max(x)), h = seq.int(0, max(x)), lty = 2, col = "grey")
matplot(x, dMat1, type = "l", ylab = "1st derivatives by 'derivs=1'")
matplot(x, dMat2, type = "l", ylab = "1st derivatives by 'deriv()'")
par(op)
### default placement of internal knots for periodic splines

default_knots <- function(x, df, intercept = FALSE,
Boundary.knots = range(x, na.rm = TRUE)) {
## get x in the cyclic interval [0, 1)
x2 <- (x - Boundary.knots[1]) %% (Boundary.knots[2] - Boundary.knots[1])
knots <- quantile(x2, probs = seq(0, 1, length.out = df + 2 - intercept))
unname(knots[- c(1, length(knots))])
}
df <- 8
degree <- 3
intercept <- TRUE
internal_knots <- default_knots(x, df, intercept)
## 1. specify df
spline_basis1 = splines2::mSpline(x, degree = degree, df = df,
periodic = TRUE, intercept = intercept)
## 2. specify knots
spline_basis2 = splines2::mSpline(x, degree = degree, knots = internal_knots,
periodic = TRUE, intercept = intercept)
all.equal(internal_knots, knots(spline_basis1))
all.equal(spline_basis1, spline_basis2)
20 naturalSpline
naturalSpline Natural Cubic Spline Basis for Polynomial Splines
Description
Generates the nonnegative natural cubic spline basis matrix, the corresponding integrals (from the
left boundary knot), or derivatives of given order. Each basis is assumed to follow a linear trend for
x outside of boundary.
Usage
naturalSpline(
x,
df = NULL,
knots = NULL,
intercept = FALSE,
derivs = 0L,
integral = FALSE,
...
)
Arguments
are.
df Degree of freedom that equals to the column number of returned matrix. One
can specify df rather than knots, then the function chooses df -1 -as.integer(intercept)
internal knots at suitable quantiles of x ignoring missing values and those x out-
side of the boundary. Thus, df must be greater than or equal to 2. If internal
knots are specified via knots, the specified df will be ignored.
derivs A nonnegative integer specifying the order of derivatives of natural splines. The
default value is 0L for the spline basis functions.
integral A logical value. The default value is FALSE. If TRUE, this function will return the
integrated natural splines from the left boundary knot.
naturalSpline 21
Details
It is an implementation of the natural spline basis based on B-spline basis, which utilizes the close-
form null space that can be derived from the recursive formula for the second derivatives of B-
splines. The constructed spline basis functions are intended to be nonnegative within boundary
with second derivatives being zeros at boundary knots.
A similar implementation is provided by splines::ns, which uses QR decomposition to find the
null space of the second derivatives of B-spline basis at boundary knots. However, there is no
guarantee that the resulting basis functions are nonnegative within boundary.
Value
A numeric matrix of length(x) rows and df columns if df is specified or length(knots) + 1 +

as.integer(intercept) columns if knots are specified instead. Attributes that correspond to the
arguments specified are returned for usage of other functions in this package.
See Also
bSpline for B-splines; mSpline for M-splines; iSpline for I-splines.
Examples
library(splines2)
x <- seq.int(0, 1, 0.01)

knots <- c(0.3, 0.5, 0.6)
## natural spline basis

nsMat0 <- naturalSpline(x, knots = knots, intercept = TRUE)
## integrals
nsMat1 <- naturalSpline(x, knots = knots, intercept = TRUE, integral = TRUE)
## first derivatives
nsMat2 <- naturalSpline(x, knots = knots, intercept = TRUE, derivs = 1)
## second derivatives
nsMat3 <- naturalSpline(x, knots = knots, intercept = TRUE, derivs = 2)
op <- par(mfrow = c(2, 2), mar = c(2.5, 2.5, 0.2, 0.1), mgp = c(1.5, 0.5, 0))
matplot(x, nsMat0, type = "l", ylab = "basis")
matplot(x, nsMat1, type = "l", ylab = "integral")
matplot(x, nsMat2, type = "l", ylab = "1st derivative")
matplot(x, nsMat3, type = "l", ylab = "2nd derivative")
par(op) # reset to previous plotting settings
## use the deriv method

all.equal(nsMat0, deriv(nsMat1), check.attributes = FALSE)
all.equal(nsMat2, deriv(nsMat0))
all.equal(nsMat3, deriv(nsMat2))
all.equal(nsMat3, deriv(nsMat0, 2))
22 predict
predict Evaluate a Spline Basis at specified points
Description
This function evaluates a predefined spline basis at a (new) given x.
Usage
predict(object, newx, ...)







Arguments
object Objects of class bSpline2, ibs, mSpline, iSpline, cSpline, bernsteinPoly
or naturalSpline with attributes describing knots, degree, etc.
newx The x values at which evaluations are required.
Details
These are methods for the generic function predict for objects inheriting from class bSpline2,
ibs, mSpline, iSpline, cSpline, naturalSpline, or bernsteinPoly. If newx is not given, the
function returns the input object.
splines2 23
Value
An object just like the object input, except evaluated at the new values of x.
Examples
library(splines2)
x <- seq.int(0, 1, 0.2)
knots <- c(0.3, 0.5, 0.6)
newX <- seq.int(0.1, 0.9, 0.2)
## for B-splines
bsMat <- bSpline(x, knots = knots, degree = 2)
predict(bsMat, newX)
## for integral of B-splines

ibsMat <- ibs(x, knots = knots, degree = 2)
predict(ibsMat, newX)
## for derivative of B-splines

dbsMat <- dbs(x, knots = knots, degree = 2)
predict(dbsMat, newX)
## for M-spline
msMat <- mSpline(x, knots = knots, degree = 2)
predict(msMat, newX)
## for I-spline
predict(isMat, newX)
## for C-spline
csMat <- cSpline(x, knots = knots, degree = 2)
predict(csMat, newX)
splines2 splines2: Regression Spline Functions and Classes
Description
This package provides functions to construct basis matrices of
• B-splines
• M-splines
• I-splines
• convex splines (C-splines)
• periodic M-splines
• natural cubic splines
24 splines2
• generalized Bernstein polynomials

• along with their integrals (except C-splines) and derivatives of given order by closed-form
recursive formulas
Details
In addition to the R interface, it also provides a C++ header-only library integrated with Rcpp,
which allows the construction of spline basis functions directly in C++ with the help of Rcpp
and RcppArmadillo. Thus, it can also be treated as one of the Rcpp* packages. A toy example
package that uses the C++ interface is available at <https://github.com/wenjie2wang/example-pkg-
Rcpp-splines2>.
The package splines2 is intended to be a user-friendly supplement to the base package splines. The
trailing number two in the package name means "too" (and by no means refers to the generation
two). See Wang and Yan (2021) for details and illustrations of how the package can be applied to
shape-restricted regression.
References
Wang, W., & Yan, J. (2021). Shape-restricted regression splines with R package splines2. Journal
of Data Science, 19(3), 498–517.
Index
bernsteinPoly, 2
bSpline, 4, 9, 13, 18, 21
cSpline, 6, 15, 18
dbs, 5, 8, 13
deriv, 10
ibs, 5, 9, 12
iSpline, 7, 13, 18, 21
knots, 15
mSpline, 7, 15, 16, 21
naturalSpline, 20
predict, 22
splines2, 23
25

Package Splines2': September 19, 2021

Uploaded by

Copyright:

Available Formats

Package Splines2': September 19, 2021

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Package Splines2': September 19, 2021

Uploaded by

Copyright:

Available Formats

Package ‘splines2’

September 19, 2021

bernsteinPoly Generalized Bernstein Polynomial Basis

x1 <- seq.int(0, 1, 0.01)

## Bernstein polynomial basis matrix over [0, 1]

## generalized Bernstein polynomials basis over [- 2, 2]

## the first and second derivative matrix

par(mfrow = c(2, 2))

## reset to previous plotting settings

## or use the deriv method

bSpline B-Spline Basis for Polynomial Splines

x <- seq.int(0, 1, 0.01)

## reset to previous plotting settings

## the first derivaitves

## the second derivaitves

## evaluate at new values

cSpline C-Spline Basis for Polynomial Splines

scale A logical value indicating if scaling C-splines is required. If TRUE by default,

x <- seq.int(0, 1, 0.01)

### when 'scale = TRUE' (by default)

## reset to previous plotting settings

### when 'scale = FALSE'

## the corresponding I-splines and M-splines (with same arguments)

isMat <- iSpline(x, knots = knots, degree = 2)

## or using deriv methods (more efficient)

dbs Derivatives of B-Splines

A numeric matrix of length(x) rows and df columns if df is specified or length(knots) + degree

bSpline for B-splines; ibs for integrals of B-splines.

## compare with the results from splineDesign

deriv Derivatives of Spline Basis Functions

## S3 method for class 'dbs'

## S3 method for class 'ibs'

## S3 method for class 'mSpline'

## S3 method for class 'iSpline'

## S3 method for class 'cSpline'

## S3 method for class 'bernsteinPoly'

## S3 method for class 'naturalSpline'

x <- c(seq.int(0, 1, 0.1), NA) # NA's will be kept.

## integal of B-splines and the corresponding B-splines integrated

## the first derivative

## the second derivative

## nested calls are supported

## C-splines, I-splines, M-splines and the derivatives

msMat <- mSpline(x, knots = knots, intercept = TRUE)

dmsMat <- mSpline(x, knots = knots, intercept = TRUE, derivs = 1)

ibs Integrals of B-Splines

A numeric matrix of length(x) rows and df columns if df is specified or length(knots) + degree

bSpline for B-splines; dbs for derivatives of B-splines;

x <- seq.int(0, 1, 0.01)

## get the corresponding B-splines by bSpline()

## plot B-spline basis with their corresponding integrals

## reset to previous plotting settings

iSpline I-Spline Basis for Polynomial Splines

## Example given in the reference paper by Ramsay (1988)

## reset to previous plotting settings

## the derivative of I-splines is M-spline

knots Extract Knots from the Given Object

## S3 method for class 'dbs'

## S3 method for class 'ibs'