ssym (version 1.5.2)

lambda.hat: Tool to choose smoothing parameters.

Description

lambda.hat is used to choose the smoothing parameter using the unweighted cross-validation score.

Usage

lambda.hat(response,xx,lambda,type,plot)

Arguments

response

the response variable.

xx

the explanatory variable.

lambda

an optional numeric value of starting estimate for the smoothing parameter. Default is 1.

type

an integer indicating the type of nonparametric effect, e.g., 1 indicates natural cubic spline, and 2 indicates P-spline.

plot

logical. If plot=TRUE, it shows a graph with the shape of the cross-validation score.

Value

• lh
  a list with the chosen smoothing parameter and the value of the cross-validation score.

References

Eilers, P.H.C. and Marx, B.D. (1996). Flexible smoothing with B-splines and penalties. Statistical Science, 11, 89-121. Green, P.J. and Silverman, B.W. (1994) Nonparametric Regression and Generalized Linear Models, Boca Raton: Chapman and Hall.

Examples

n <- 300
t <- sort(round(runif(n),digits=2))
y <- cos(4*pi*t) + rnorm(n)
datas <- data.frame(t,y)

par(mfrow=c(1,2))
lambda <- lambda.hat(y,t,1,1,plot=TRUE)
lambda <- lambda$lambda_hat
fit <- ssym.l(y ~ ncs(t, lambda=lambda), family="Normal", data=datas)
np.graph(fit, which=1, main="Natural Cubic Spline")

#x11()
par(mfrow=c(1,2))
lambda <- lambda.hat(y,t,1,2,plot=TRUE)
lambda <- lambda$lambda_hat
fit2 <- ssym.l(y ~ psp(t, lambda=lambda), family="Normal", data=datas)
np.graph(fit2, which=1, main="P-spline")