Get predictions from a GAM model.

Source: R/utils.R

It returns a tibble with the predictions from a gam or bam object.

get_gam_predictions(
  model,
  series,
  series_length = 25,
  conditions = NULL,
  exclude_random = TRUE,
  exclude_terms = NULL,
  split = NULL,
  sep = "\\.",
  time_series,
  transform = NULL,
  ci_z = 1.96,
  .comparison = NULL
)

Arguments

model

A gam or bam model object.
series

An unquoted expression indicating the model term that defines the series on which smoothing is applied. This is the term that is displayed on the x-axis when plotting.
series_length

An integer indicating how many values along the time series to use for predicting the outcome term.
conditions

A list of quosures with quos specifying the levels to plot from the model terms.
exclude_random

Whether to exclude random smooths (the default is TRUE).
exclude_terms

Terms to be excluded from the prediction. Term names should be given as they appear in the model summary (for example, "s(x0,x1)").
split

Columns to separate as a named list.
sep

Separator between columns (default is "\.", which is the default with ). If character, it is interpreted as a regular expression.
time_series

Deprecated, use series instead.
transform

Function used to transform the fitted values (useful for getting plots on the response scale).
ci_z

The z-value for calculating the CIs (the default is 1.96 for 95 percent CI).
.comparison

Internal parameter, passed from plot_smooths().

Value

A tibble with predictions from a gam or bam model.

Examples

library(mgcv)
set.seed(10)
data <- gamSim(4)

#> Factor `by' variable example
model <- gam(y ~ fac + s(x2) + s(x2, by = fac) + s(x0), data = data)

pred <- get_gam_predictions(model, x2)