sitar (version 1.5.0)

predict.sitar: Predict SITAR model

Description

Predict method for sitar objects, based on predict.lme.

Usage

# S3 method for sitar
predict(
  object,
  newdata = getData(object),
  level = 1L,
  ...,
  deriv = 0L,
  abc = NULL
)

Value

A vector of the predictions, or a list of vectors if asList = TRUE and level == 1, or a data frame if length(level) * length(deriv) > 1. The data frame column names are (a subset of): (id, predict.fixed, predict.id, vel.predict.fixed, vel.predict.id).

Arguments

object

an object inheriting from class sitar.

newdata

an optional data frame to be used for obtaining the predictions, defaulting to the data used to fit object. It requires named columns for x, and also for id if level = 1, matching the names in object. Variables with the reserved names x = .x or id = .id take precedence over the model x and id variables. Any covariates in a.formula, b.formula, c.formula or d.formula can also be included. By default their values are set to the mean, so when level = 0 the prediction represents the mean curve.

level

an optional integer vector giving the level(s) of grouping to be used in obtaining the predictions, level 0 corresponding to the population predictions. Defaults to level 1, and level = 0:1 fits both levels.

...

other optional arguments: asList, na.action and naPattern.

deriv

an optional integer specifying predictions corresponding to the fitted curve and/or its derivative. deriv = 0 (default) specifies the distance curve, deriv = 1 the velocity curve and deriv = 0:1 fits both.

abc

an optional named vector containing values of a subset of a, b, c and d, default NULL. Ignored if level = 0. It gives predictions for a single subject with the specified values of a, b, c and d, where missing values are set to 0. Alternatively abc can contain the value for a single id.

Author

Tim Cole

Details

If x and/or y include transformations, e.g. x = log(age), predictions are returned in the original units by back-transforming x and/or y appropriately using the function ifun.

Examples

data(heights)
require(dplyr)
require(tidyr)
require(ggplot2)
## fit model
m1 <- sitar(x = age, y = height, id = id, data = heights, df = 5)
df <- data.frame(age = 9:16, id = 5)

## height predictions at level 0
predict(m1, newdata = df, level = 0)

## height predictions at level 1 for subject 5
predict(m1, newdata = df, level = 1)

## velocity predictions for subjects with early, average and late puberty
m2 <- sitar(x = log(age), y = height, id = id, data = heights, df = 5)
tibble(age = 80:160/10) %>%
  mutate(early = predict(m2, ., deriv = 1, abc = c(b = -0.07)),
         average = predict(m2, ., deriv = 1, level = 0),
         late = predict(m2, ., deriv = 1, abc = c(b = 0.07))) %>%
  pivot_longer(early:late, names_to = 'group', values_to = 'height_velocity') %>%
  ggplot(aes(age, height_velocity, group = group, colour = group)) +
  geom_path(show.legend = FALSE)