The first derivative of the smooth functions of a GAM model calculated using finite differences.

fderiv(model, ...)

# S3 method for gam
fderiv(model, newdata, term, n = 200, eps = 1e-07,
  unconditional = FALSE, offset = NULL, ...)

# S3 method for gamm
fderiv(model, ...)

Arguments

model

A fitted GAM. Currently only models fitted by mgcv::gam() and mgcv::gamm() are supported.

...

Arguments that are passed to other methods.

newdata

a data frame containing the values of the model covariates at which to evaluate the first derivatives of the smooths.

term

character; vector of one or more terms for which derivatives are required. If missing, derivatives for all smooth terms will be returned.

n

integer; if newdata is missing the original data can be reconstructed from model and then n controls the number of values over the range of each covariate with which to populate newdata.

eps

numeric; the value of the finite difference used to approximate the first derivative.

unconditional

logical; if TRUE, the smoothing parameter uncertainty corrected covariance matrix is used, if available, otherwise the uncorrected Bayesian posterior covariance matrix is used.

offset

numeric; value of offset to use in generating predictions.

Value

An object of class "fderiv" is returned.

Examples

dat <- gamSim(1, n = 400, dist = "normal", scale = 2)
#> Gu & Wahba 4 term additive model
mod <- gam(y ~ s(x0) + s(x1) + s(x2) + s(x3), data = dat, method = "REML") ## first derivatives of all smooths... fd <- fderiv(mod) ## ...and a selected smooth fd2 <- fderiv(mod, term = "x1") ## Models with factors set.seed(2) dat <- gamSim(4, n = 400, dist = "normal", scale = 2)
#> Factor `by' variable example
mod <- gam(y ~ s(x0) + s(x1) + fac, data = dat, method = "REML") ## first derivatives of all smooths... fd <- fderiv(mod) ## ...and a selected smooth fd2 <- fderiv(mod, term = "x1")