mcp: get_segment_table – R documentation

Become an expert in R — Interactive courses, Cheat Sheets, certificates and more!

get_segment_table

Build a table describing a list of segments

Used internally for most mcp functions.

get_segment_table(model, data = NULL, family = gaussian(), par_x = NULL)

`model`	A list of formulas - one for each segment. The first formula has the format `response ~ predictors` while the following formulas have the format `response ~ changepoint ~ predictors`. The response and change points can be omitted (`changepoint ~ predictors` assumes same response. `~ predictors` assumes an intercept-only change point). The following can be modeled: Regular formulas: e.g., `~ 1 + x`). Read more. Extended formulas:, e.g., `~ I(x^2) + exp(x) + sin(x)`. Read more. Variance: e.g., `~sigma(1)` for a simple variance change or `~sigma(rel(1) + I(x^2))`) for more advanced variance structures. Read more Autoregression: e.g., `~ar(1)` for a simple onset/change in AR(1) or `ar(2, 0 + x`) for an AR(2) increasing by `x`. Read more
`data`	Data.frame or tibble in long format.
`family`	One of `gaussian()`, `binomial()`, `bernoulli()`, or `poission()`. Only default link functions are currently supported.
`par_x`	String (default: NULL). Only relevant if no segments contains slope (no hint at what x is). Set this, e.g., par_x = "time".

A tibble with one row describing each segment and the corresponding code.

Jonas Kristoffer Lindeløv jonas@lindeloev.dk

model = list(
  y ~ 1 + x,
  1 + (1|id) ~ 1
)
get_segment_table(model)

Regression with Multiple Change Points

v0.3.0

GPL-2

Authors

Jonas Kristoffer Lindeløv [aut, cre] (<https://orcid.org/0000-0003-4565-0595>)

Initial release

2020-07-27