Package 'broom'

Description

augment_columns is intended for use in the internals of augment methods only and is exported for developers extending the broom package. Please instead use augment() to appropriately make use of the functionality in augment_columns().

Usage

augment_columns(
  x,
  data,
  newdata = NULL,
  type,
  type.predict = type,
  type.residuals = type,
  se.fit = TRUE,
  ...
)

Arguments

Details

Note that, in the case that a residuals() or influence() generic is not implemented for the supplied model x, the function will fail quietly.

Description

Augment accepts a model object and a dataset and adds information about each observation in the dataset. Most commonly, this includes predicted values in the .fitted column, residuals in the .resid column, and standard errors for the fitted values in a .se.fit column. New columns always begin with a . prefix to avoid overwriting columns in the original dataset.

Users may pass data to augment via either the data argument or the newdata argument. If the user passes data to the data argument, it must be exactly the data that was used to fit the model object. Pass datasets to newdata to augment data that was not used during model fitting. This still requires that at least all predictor variable columns used to fit the model are present. If the original outcome variable used to fit the model is not included in newdata, then no .resid column will be included in the output.

Augment will often behave differently depending on whether data or newdata is given. This is because there is often information associated with training observations (such as influences or related) measures that is not meaningfully defined for new observations.

For convenience, many augment methods provide default data arguments, so that augment(fit) will return the augmented training data. In these cases, augment tries to reconstruct the original data based on the model object with varying degrees of success.

The augmented dataset is always returned as a tibble::tibble with the same number of rows as the passed dataset. This means that the passed data must be coercible to a tibble. If a predictor enters the model as part of a matrix of covariates, such as when the model formula uses splines::ns(), stats::poly(), or survival::Surv(), it is represented as a matrix column.

We are in the process of defining behaviors for models fit with various na.action arguments, but make no guarantees about behavior when data is missing at this time.

Usage

## S3 method for class 'betamfx'
augment(
  x,
  data = model.frame(x$fit),
  newdata = NULL,
  type.predict = c("response", "link", "precision", "variance", "quantile"),
  type.residuals = c("sweighted2", "deviance", "pearson", "response", "weighted",
    "sweighted"),
  ...
)

Arguments

Details

This augment method wraps augment.betareg() for mfx::betamfx() objects.

Value

A tibble::tibble() with columns:

See Also

augment.betareg(), mfx::betamfx()

Other mfx tidiers: augment.mfx(), glance.betamfx(), glance.mfx(), tidy.betamfx(), tidy.mfx()

Examples

library(mfx)

# Simulate some data
set.seed(12345)
n <- 1000
x <- rnorm(n)

# Beta outcome
y <- rbeta(n, shape1 = plogis(1 + 0.5 * x), shape2 = (abs(0.2 * x)))
# Use Smithson and Verkuilen correction
y <- (y * (n - 1) + 0.5) / n

d <- data.frame(y, x)
mod_betamfx <- betamfx(y ~ x | x, data = d)

tidy(mod_betamfx, conf.int = TRUE)

# Compare with the naive model coefficients of the equivalent betareg call (not run)
# tidy(betamfx(y ~ x | x, data = d), conf.int = TRUE)

augment(mod_betamfx)
glance(mod_betamfx)

Description

Augment accepts a model object and a dataset and adds information about each observation in the dataset. Most commonly, this includes predicted values in the .fitted column, residuals in the .resid column, and standard errors for the fitted values in a .se.fit column. New columns always begin with a . prefix to avoid overwriting columns in the original dataset.

Users may pass data to augment via either the data argument or the newdata argument. If the user passes data to the data argument, it must be exactly the data that was used to fit the model object. Pass datasets to newdata to augment data that was not used during model fitting. This still requires that at least all predictor variable columns used to fit the model are present. If the original outcome variable used to fit the model is not included in newdata, then no .resid column will be included in the output.

Augment will often behave differently depending on whether data or newdata is given. This is because there is often information associated with training observations (such as influences or related) measures that is not meaningfully defined for new observations.

For convenience, many augment methods provide default data arguments, so that augment(fit) will return the augmented training data. In these cases, augment tries to reconstruct the original data based on the model object with varying degrees of success.

The augmented dataset is always returned as a tibble::tibble with the same number of rows as the passed dataset. This means that the passed data must be coercible to a tibble. If a predictor enters the model as part of a matrix of covariates, such as when the model formula uses splines::ns(), stats::poly(), or survival::Surv(), it is represented as a matrix column.

We are in the process of defining behaviors for models fit with various na.action arguments, but make no guarantees about behavior when data is missing at this time.

Usage

## S3 method for class 'betareg'
augment(
  x,
  data = model.frame(x),
  newdata = NULL,
  type.predict,
  type.residuals,
  ...
)

Arguments

Details

For additional details on Cook's distance, see stats::cooks.distance().

Value

A tibble::tibble() with columns:

See Also

augment(), betareg::betareg()

Examples

# load libraries for models and data
library(betareg)

# load dats
data("GasolineYield", package = "betareg")

# fit model
mod <- betareg(yield ~ batch + temp, data = GasolineYield)

mod

# summarize model fit with tidiers
tidy(mod)
tidy(mod, conf.int = TRUE)
tidy(mod, conf.int = TRUE, conf.level = .99)

augment(mod)

glance(mod)

Description

Augment accepts a model object and a dataset and adds information about each observation in the dataset. Most commonly, this includes predicted values in the .fitted column, residuals in the .resid column, and standard errors for the fitted values in a .se.fit column. New columns always begin with a . prefix to avoid overwriting columns in the original dataset.

Users may pass data to augment via either the data argument or the newdata argument. If the user passes data to the data argument, it must be exactly the data that was used to fit the model object. Pass datasets to newdata to augment data that was not used during model fitting. This still requires that at least all predictor variable columns used to fit the model are present. If the original outcome variable used to fit the model is not included in newdata, then no .resid column will be included in the output.

Augment will often behave differently depending on whether data or newdata is given. This is because there is often information associated with training observations (such as influences or related) measures that is not meaningfully defined for new observations.

For convenience, many augment methods provide default data arguments, so that augment(fit) will return the augmented training data. In these cases, augment tries to reconstruct the original data based on the model object with varying degrees of success.

The augmented dataset is always returned as a tibble::tibble with the same number of rows as the passed dataset. This means that the passed data must be coercible to a tibble. If a predictor enters the model as part of a matrix of covariates, such as when the model formula uses splines::ns(), stats::poly(), or survival::Surv(), it is represented as a matrix column.

We are in the process of defining behaviors for models fit with various na.action arguments, but make no guarantees about behavior when data is missing at this time.

Usage

## S3 method for class 'clm'
augment(
  x,
  data = model.frame(x),
  newdata = NULL,
  type.predict = c("prob", "class"),
  ...
)

Arguments

See Also

tidy, ordinal::clm(), ordinal::predict.clm()

Other ordinal tidiers: augment.polr(), glance.clm(), glance.clmm(), glance.polr(), glance.svyolr(), tidy.clm(), tidy.clmm(), tidy.polr(), tidy.svyolr()

Examples

# load libraries for models and data
library(ordinal)

# fit model
fit <- clm(rating ~ temp * contact, data = wine)

# summarize model fit with tidiers
tidy(fit)
tidy(fit, conf.int = TRUE, conf.level = 0.9)
tidy(fit, conf.int = TRUE, conf.type = "Wald", exponentiate = TRUE)

glance(fit)
augment(fit, type.predict = "prob")
augment(fit, type.predict = "class")

# ...and again with another model specification
fit2 <- clm(rating ~ temp, nominal = ~contact, data = wine)

tidy(fit2)
glance(fit2)

Description

Augment accepts a model object and a dataset and adds information about each observation in the dataset. Most commonly, this includes predicted values in the .fitted column, residuals in the .resid column, and standard errors for the fitted values in a .se.fit column. New columns always begin with a . prefix to avoid overwriting columns in the original dataset.

Users may pass data to augment via either the data argument or the newdata argument. If the user passes data to the data argument, it must be exactly the data that was used to fit the model object. Pass datasets to newdata to augment data that was not used during model fitting. This still requires that at least all predictor variable columns used to fit the model are present. If the original outcome variable used to fit the model is not included in newdata, then no .resid column will be included in the output.

Augment will often behave differently depending on whether data or newdata is given. This is because there is often information associated with training observations (such as influences or related) measures that is not meaningfully defined for new observations.

For convenience, many augment methods provide default data arguments, so that augment(fit) will return the augmented training data. In these cases, augment tries to reconstruct the original data based on the model object with varying degrees of success.

The augmented dataset is always returned as a tibble::tibble with the same number of rows as the passed dataset. This means that the passed data must be coercible to a tibble. If a predictor enters the model as part of a matrix of covariates, such as when the model formula uses splines::ns(), stats::poly(), or survival::Surv(), it is represented as a matrix column.

We are in the process of defining behaviors for models fit with various na.action arguments, but make no guarantees about behavior when data is missing at this time.

Usage

## S3 method for class 'coxph'
augment(
  x,
  data = model.frame(x),
  newdata = NULL,
  type.predict = "lp",
  type.residuals = "martingale",
  ...
)

Arguments

Details

When the modeling was performed with na.action = "na.omit" (as is the typical default), rows with NA in the initial data are omitted entirely from the augmented data frame. When the modeling was performed with na.action = "na.exclude", one should provide the original data as a second argument, at which point the augmented data will contain those rows (typically with NAs in place of the new columns). If the original data is not provided to augment() and na.action = "na.exclude", a warning is raised and the incomplete rows are dropped.

Value

A tibble::tibble() with columns:

See Also

stats::na.action

augment(), survival::coxph()

Other coxph tidiers: glance.coxph(), tidy.coxph()

Other survival tidiers: augment.survreg(), glance.aareg(), glance.cch(), glance.coxph(), glance.pyears(), glance.survdiff(), glance.survexp(), glance.survfit(), glance.survreg(), tidy.aareg(), tidy.cch(), tidy.coxph(), tidy.pyears(), tidy.survdiff(), tidy.survexp(), tidy.survfit(), tidy.survreg()

Examples

# load libraries for models and data
library(survival)

# fit model
cfit <- coxph(Surv(time, status) ~ age + sex, lung)

# summarize model fit with tidiers
tidy(cfit)
tidy(cfit, exponentiate = TRUE)

lp <- augment(cfit, lung)
risks <- augment(cfit, lung, type.predict = "risk")
expected <- augment(cfit, lung, type.predict = "expected")

glance(cfit)

# also works on clogit models
resp <- levels(logan$occupation)
n <- nrow(logan)
indx <- rep(1:n, length(resp))
logan2 <- data.frame(
  logan[indx, ],
  id = indx,
  tocc = factor(rep(resp, each = n))
)

logan2$case <- (logan2$occupation == logan2$tocc)

cl <- clogit(case ~ tocc + tocc:education + strata(id), logan2)

tidy(cl)
glance(cl)

library(ggplot2)

ggplot(lp, aes(age, .fitted, color = sex)) +
  geom_point()

ggplot(risks, aes(age, .fitted, color = sex)) +
  geom_point()

ggplot(expected, aes(time, .fitted, color = sex)) +
  geom_point()

Description

Augment accepts a model object and a dataset and adds information about each observation in the dataset. Most commonly, this includes predicted values in the .fitted column, residuals in the .resid column, and standard errors for the fitted values in a .se.fit column. New columns always begin with a . prefix to avoid overwriting columns in the original dataset.

Users may pass data to augment via either the data argument or the newdata argument. If the user passes data to the data argument, it must be exactly the data that was used to fit the model object. Pass datasets to newdata to augment data that was not used during model fitting. This still requires that at least all predictor variable columns used to fit the model are present. If the original outcome variable used to fit the model is not included in newdata, then no .resid column will be included in the output.

Augment will often behave differently depending on whether data or newdata is given. This is because there is often information associated with training observations (such as influences or related) measures that is not meaningfully defined for new observations.

For convenience, many augment methods provide default data arguments, so that augment(fit) will return the augmented training data. In these cases, augment tries to reconstruct the original data based on the model object with varying degrees of success.

The augmented dataset is always returned as a tibble::tibble with the same number of rows as the passed dataset. This means that the passed data must be coercible to a tibble. If a predictor enters the model as part of a matrix of covariates, such as when the model formula uses splines::ns(), stats::poly(), or survival::Surv(), it is represented as a matrix column.

We are in the process of defining behaviors for models fit with various na.action arguments, but make no guarantees about behavior when data is missing at this time.

Usage

## S3 method for class 'decomposed.ts'
augment(x, ...)

Arguments

Value

A tibble::tibble with one row for each observation in the original times series:

See Also

augment(), stats::decompose()

Other decompose tidiers: augment.stl()

Examples

# time series of temperatures in Nottingham, 1920-1939:
nottem

# perform seasonal decomposition on the data with both decompose
# and stl:
d1 <- decompose(nottem)
d2 <- stl(nottem, s.window = "periodic", robust = TRUE)

# compare the original series to its decompositions.

cbind(
  tidy(nottem), augment(d1),
  augment(d2)
)

# visually compare seasonal decompositions in tidy data frames.

library(tibble)
library(dplyr)
library(tidyr)
library(ggplot2)

decomps <- tibble(
  # turn the ts objects into data frames.
  series = list(as.data.frame(nottem), as.data.frame(nottem)),
  # add the models in, one for each row.
  decomp = c("decompose", "stl"),
  model = list(d1, d2)
) %>%
  rowwise() %>%
  # pull out the fitted data using broom::augment.
  mutate(augment = list(broom::augment(model))) %>%
  ungroup() %>%
  # unnest the data frames into a tidy arrangement of
  # the series next to its seasonal decomposition, grouped
  # by the method (stl or decompose).
  group_by(decomp) %>%
  unnest(c(series, augment)) %>%
  mutate(index = 1:n()) %>%
  ungroup() %>%
  select(decomp, index, x, adjusted = .seasadj)

ggplot(decomps) +
  geom_line(aes(x = index, y = x), colour = "black") +
  geom_line(aes(
    x = index, y = adjusted, colour = decomp,
    group = decomp
  ))

Description

Augment accepts a model object and a dataset and adds information about each observation in the dataset. Most commonly, this includes predicted values in the .fitted column, residuals in the .resid column, and standard errors for the fitted values in a .se.fit column. New columns always begin with a . prefix to avoid overwriting columns in the original dataset.

Users may pass data to augment via either the data argument or the newdata argument. If the user passes data to the data argument, it must be exactly the data that was used to fit the model object. Pass datasets to newdata to augment data that was not used during model fitting. This still requires that at least all predictor variable columns used to fit the model are present. If the original outcome variable used to fit the model is not included in newdata, then no .resid column will be included in the output.

Augment will often behave differently depending on whether data or newdata is given. This is because there is often information associated with training observations (such as influences or related) measures that is not meaningfully defined for new observations.

For convenience, many augment methods provide default data arguments, so that augment(fit) will return the augmented training data. In these cases, augment tries to reconstruct the original data based on the model object with varying degrees of success.

The augmented dataset is always returned as a tibble::tibble with the same number of rows as the passed dataset. This means that the passed data must be coercible to a tibble. If a predictor enters the model as part of a matrix of covariates, such as when the model formula uses splines::ns(), stats::poly(), or survival::Surv(), it is represented as a matrix column.

We are in the process of defining behaviors for models fit with various na.action arguments, but make no guarantees about behavior when data is missing at this time.

Usage

## S3 method for class 'drc'
augment(
  x,
  data = NULL,
  newdata = NULL,
  se_fit = FALSE,
  conf.int = FALSE,
  conf.level = 0.95,
  ...
)

Arguments

Value

A tibble::tibble() with columns:

See Also

augment(), drc::drm()

Other drc tidiers: glance.drc(), tidy.drc()

Examples

# load libraries for models and data
library(drc)

# fit model
mod <- drm(dead / total ~ conc, type,
  weights = total, data = selenium, fct = LL.2(), type = "binomial"
)

# summarize model fit with tidiers
tidy(mod)
tidy(mod, conf.int = TRUE)

glance(mod)

augment(mod, selenium)

Description

Augment accepts a model object and a dataset and adds information about each observation in the dataset. Most commonly, this includes predicted values in the .fitted column, residuals in the .resid column, and standard errors for the fitted values in a .se.fit column. New columns always begin with a . prefix to avoid overwriting columns in the original dataset.

Users may pass data to augment via either the data argument or the newdata argument. If the user passes data to the data argument, it must be exactly the data that was used to fit the model object. Pass datasets to newdata to augment data that was not used during model fitting. This still requires that at least all predictor variable columns used to fit the model are present. If the original outcome variable used to fit the model is not included in newdata, then no .resid column will be included in the output.

Augment will often behave differently depending on whether data or newdata is given. This is because there is often information associated with training observations (such as influences or related) measures that is not meaningfully defined for new observations.

For convenience, many augment methods provide default data arguments, so that augment(fit) will return the augmented training data. In these cases, augment tries to reconstruct the original data based on the model object with varying degrees of success.

The augmented dataset is always returned as a tibble::tibble with the same number of rows as the passed dataset. This means that the passed data must be coercible to a tibble. If a predictor enters the model as part of a matrix of covariates, such as when the model formula uses splines::ns(), stats::poly(), or survival::Surv(), it is represented as a matrix column.

We are in the process of defining behaviors for models fit with various na.action arguments, but make no guarantees about behavior when data is missing at this time.

Usage

## S3 method for class 'factanal'
augment(x, data, ...)

Arguments

Value

When data is not supplied augment.factanal returns one row for each observation, with a factor score column added for each factor X, (.fsX). This is because stats::factanal(), unlike other stats methods like stats::lm(), does not retain the original data.

When data is supplied, augment.factanal returns one row for each observation, with a factor score column added for each factor X, (.fsX).

See Also

augment(), stats::factanal()

Other factanal tidiers: glance.factanal(), tidy.factanal()

Description

Augment accepts a model object and a dataset and adds information about each observation in the dataset. Most commonly, this includes predicted values in the .fitted column, residuals in the .resid column, and standard errors for the fitted values in a .se.fit column. New columns always begin with a . prefix to avoid overwriting columns in the original dataset.

Users may pass data to augment via either the data argument or the newdata argument. If the user passes data to the data argument, it must be exactly the data that was used to fit the model object. Pass datasets to newdata to augment data that was not used during model fitting. This still requires that at least all predictor variable columns used to fit the model are present. If the original outcome variable used to fit the model is not included in newdata, then no .resid column will be included in the output.

Augment will often behave differently depending on whether data or newdata is given. This is because there is often information associated with training observations (such as influences or related) measures that is not meaningfully defined for new observations.

For convenience, many augment methods provide default data arguments, so that augment(fit) will return the augmented training data. In these cases, augment tries to reconstruct the original data based on the model object with varying degrees of success.

The augmented dataset is always returned as a tibble::tibble with the same number of rows as the passed dataset. This means that the passed data must be coercible to a tibble. If a predictor enters the model as part of a matrix of covariates, such as when the model formula uses splines::ns(), stats::poly(), or survival::Surv(), it is represented as a matrix column.

We are in the process of defining behaviors for models fit with various na.action arguments, but make no guarantees about behavior when data is missing at this time.

Usage

## S3 method for class 'felm'
augment(x, data = model.frame(x), ...)

Arguments

Value

A tibble::tibble() with columns:

See Also

augment(), lfe::felm()

Other felm tidiers: tidy.felm()

Examples

# load libraries for models and data
library(lfe)

# use built-in `airquality` dataset
head(airquality)

# no FEs; same as lm()
est0 <- felm(Ozone ~ Temp + Wind + Solar.R, airquality)

# summarize model fit with tidiers
tidy(est0)
augment(est0)

# add month fixed effects
est1 <- felm(Ozone ~ Temp + Wind + Solar.R | Month, airquality)

# summarize model fit with tidiers
tidy(est1)
tidy(est1, fe = TRUE)
augment(est1)
glance(est1)

# the "se.type" argument can be used to switch out different standard errors
# types on the fly. In turn, this can be useful exploring the effect of
# different error structures on model inference.
tidy(est1, se.type = "iid")
tidy(est1, se.type = "robust")

# add clustered SEs (also by month)
est2 <- felm(Ozone ~ Temp + Wind + Solar.R | Month | 0 | Month, airquality)

# summarize model fit with tidiers
tidy(est2, conf.int = TRUE)
tidy(est2, conf.int = TRUE, se.type = "cluster")
tidy(est2, conf.int = TRUE, se.type = "robust")
tidy(est2, conf.int = TRUE, se.type = "iid")

Description

Augment accepts a model object and a dataset and adds information about each observation in the dataset. Most commonly, this includes predicted values in the .fitted column, residuals in the .resid column, and standard errors for the fitted values in a .se.fit column. New columns always begin with a . prefix to avoid overwriting columns in the original dataset.

Users may pass data to augment via either the data argument or the newdata argument. If the user passes data to the data argument, it must be exactly the data that was used to fit the model object. Pass datasets to newdata to augment data that was not used during model fitting. This still requires that at least all predictor variable columns used to fit the model are present. If the original outcome variable used to fit the model is not included in newdata, then no .resid column will be included in the output.

Augment will often behave differently depending on whether data or newdata is given. This is because there is often information associated with training observations (such as influences or related) measures that is not meaningfully defined for new observations.

For convenience, many augment methods provide default data arguments, so that augment(fit) will return the augmented training data. In these cases, augment tries to reconstruct the original data based on the model object with varying degrees of success.

The augmented dataset is always returned as a tibble::tibble with the same number of rows as the passed dataset. This means that the passed data must be coercible to a tibble. If a predictor enters the model as part of a matrix of covariates, such as when the model formula uses splines::ns(), stats::poly(), or survival::Surv(), it is represented as a matrix column.

We are in the process of defining behaviors for models fit with various na.action arguments, but make no guarantees about behavior when data is missing at this time.

Usage

## S3 method for class 'fixest'
augment(
  x,
  data = NULL,
  newdata = NULL,
  type.predict = c("link", "response"),
  type.residuals = c("response", "deviance", "pearson", "working"),
  ...
)

Arguments

Value

A tibble::tibble() with columns:

Note

Important note: fixest models do not include a copy of the input data, so you must provide it manually.

augment.fixest only works for fixest::feols(), fixest::feglm(), and fixest::femlm() models. It does not work with results from fixest::fenegbin(), fixest::feNmlm(), or fixest::fepois().

See Also

augment(), fixest::feglm(), fixest::femlm(), fixest::feols()

Other fixest tidiers: tidy.fixest()

Examples

# load libraries for models and data
library(fixest)

gravity <-
  feols(
    log(Euros) ~ log(dist_km) | Origin + Destination + Product + Year, trade
  )

tidy(gravity)
glance(gravity)
augment(gravity, trade)

# to get robust or clustered SEs, users can either:

# 1) specify the arguments directly in the `tidy()` call

tidy(gravity, conf.int = TRUE, cluster = c("Product", "Year"))

tidy(gravity, conf.int = TRUE, se = "threeway")

# 2) or, feed tidy() a summary.fixest object that has already accepted
# these arguments

gravity_summ <- summary(gravity, cluster = c("Product", "Year"))

tidy(gravity_summ, conf.int = TRUE)

# approach (1) is preferred.

Description

Augment accepts a model object and a dataset and adds information about each observation in the dataset. Most commonly, this includes predicted values in the .fitted column, residuals in the .resid column, and standard errors for the fitted values in a .se.fit column. New columns always begin with a . prefix to avoid overwriting columns in the original dataset.

Users may pass data to augment via either the data argument or the newdata argument. If the user passes data to the data argument, it must be exactly the data that was used to fit the model object. Pass datasets to newdata to augment data that was not used during model fitting. This still requires that at least all predictor variable columns used to fit the model are present. If the original outcome variable used to fit the model is not included in newdata, then no .resid column will be included in the output.

Augment will often behave differently depending on whether data or newdata is given. This is because there is often information associated with training observations (such as influences or related) measures that is not meaningfully defined for new observations.

For convenience, many augment methods provide default data arguments, so that augment(fit) will return the augmented training data. In these cases, augment tries to reconstruct the original data based on the model object with varying degrees of success.

The augmented dataset is always returned as a tibble::tibble with the same number of rows as the passed dataset. This means that the passed data must be coercible to a tibble. If a predictor enters the model as part of a matrix of covariates, such as when the model formula uses splines::ns(), stats::poly(), or survival::Surv(), it is represented as a matrix column.

We are in the process of defining behaviors for models fit with various na.action arguments, but make no guarantees about behavior when data is missing at this time.

Usage

## S3 method for class 'gam'
augment(
  x,
  data = model.frame(x),
  newdata = NULL,
  type.predict,
  type.residuals,
  ...
)

Arguments

Details

For additional details on Cook's distance, see stats::cooks.distance().

Value

A tibble::tibble() with columns:

See Also

augment(), mgcv::gam()

Examples

# load libraries for models and data
library(mgcv)

# fit model
g <- gam(mpg ~ s(hp) + am + qsec, data = mtcars)

# summarize model fit with tidiers
tidy(g)
tidy(g, parametric = TRUE)
glance(g)
augment(g)

Description

Augment accepts a model object and a dataset and adds information about each observation in the dataset. Most commonly, this includes predicted values in the .fitted column, residuals in the .resid column, and standard errors for the fitted values in a .se.fit column. New columns always begin with a . prefix to avoid overwriting columns in the original dataset.

Users may pass data to augment via either the data argument or the newdata argument. If the user passes data to the data argument, it must be exactly the data that was used to fit the model object. Pass datasets to newdata to augment data that was not used during model fitting. This still requires that at least all predictor variable columns used to fit the model are present. If the original outcome variable used to fit the model is not included in newdata, then no .resid column will be included in the output.

Augment will often behave differently depending on whether data or newdata is given. This is because there is often information associated with training observations (such as influences or related) measures that is not meaningfully defined for new observations.

For convenience, many augment methods provide default data arguments, so that augment(fit) will return the augmented training data. In these cases, augment tries to reconstruct the original data based on the model object with varying degrees of success.

The augmented dataset is always returned as a tibble::tibble with the same number of rows as the passed dataset. This means that the passed data must be coercible to a tibble. If a predictor enters the model as part of a matrix of covariates, such as when the model formula uses splines::ns(), stats::poly(), or survival::Surv(), it is represented as a matrix column.

We are in the process of defining behaviors for models fit with various na.action arguments, but make no guarantees about behavior when data is missing at this time.

Usage

## S3 method for class 'glm'
augment(
  x,
  data = model.frame(x),
  newdata = NULL,
  type.predict = c("link", "response", "terms"),
  type.residuals = c("deviance", "pearson"),
  se_fit = FALSE,
  ...
)

Arguments

Details

If the weights for any of the observations in the model are 0, then columns ".infl" and ".hat" in the result will be 0 for those observations.

A .resid column is not calculated when data is specified via the newdata argument.

Value

A tibble::tibble() with columns:

See Also

stats::glm()

Other lm tidiers: augment.lm(), glance.glm(), glance.lm(), glance.summary.lm(), glance.svyglm(), tidy.glm(), tidy.lm(), tidy.lm.beta(), tidy.mlm(), tidy.summary.lm()

Description

Augment accepts a model object and a dataset and adds information about each observation in the dataset. Most commonly, this includes predicted values in the .fitted column, residuals in the .resid column, and standard errors for the fitted values in a .se.fit column. New columns always begin with a . prefix to avoid overwriting columns in the original dataset.

Users may pass data to augment via either the data argument or the newdata argument. If the user passes data to the data argument, it must be exactly the data that was used to fit the model object. Pass datasets to newdata to augment data that was not used during model fitting. This still requires that at least all predictor variable columns used to fit the model are present. If the original outcome variable used to fit the model is not included in newdata, then no .resid column will be included in the output.

Augment will often behave differently depending on whether data or newdata is given. This is because there is often information associated with training observations (such as influences or related) measures that is not meaningfully defined for new observations.

For convenience, many augment methods provide default data arguments, so that augment(fit) will return the augmented training data. In these cases, augment tries to reconstruct the original data based on the model object with varying degrees of success.

The augmented dataset is always returned as a tibble::tibble with the same number of rows as the passed dataset. This means that the passed data must be coercible to a tibble. If a predictor enters the model as part of a matrix of covariates, such as when the model formula uses splines::ns(), stats::poly(), or survival::Surv(), it is represented as a matrix column.

We are in the process of defining behaviors for models fit with various na.action arguments, but make no guarantees about behavior when data is missing at this time.

Usage

## S3 method for class 'glmrob'
augment(
  x,
  data = model.frame(x),
  newdata = NULL,
  type.predict = c("link", "response"),
  type.residuals = c("deviance", "pearson"),
  se_fit = FALSE,
  ...
)

Arguments

Details

For tidiers for robust models from the MASS package see tidy.rlm().

Value

A tibble::tibble() with columns:

See Also

robustbase::glmrob()

Other robustbase tidiers: augment.lmrob(), glance.lmrob(), tidy.glmrob(), tidy.lmrob()

Examples

if (requireNamespace("robustbase", quietly = TRUE)) {
  # load libraries for models and data
  library(robustbase)

  data(coleman)
  set.seed(0)

  m <- lmrob(Y ~ ., data = coleman)
  tidy(m)
  augment(m)
  glance(m)

  data(carrots)

  Rfit <- glmrob(cbind(success, total - success) ~ logdose + block,
    family = binomial, data = carrots, method = "Mqle",
    control = glmrobMqle.control(tcc = 1.2)
  )

  tidy(Rfit)
  augment(Rfit)
}

Description

Augment accepts a model object and a dataset and adds information about each observation in the dataset. Most commonly, this includes predicted values in the .fitted column, residuals in the .resid column, and standard errors for the fitted values in a .se.fit column. New columns always begin with a . prefix to avoid overwriting columns in the original dataset.

Users may pass data to augment via either the data argument or the newdata argument. If the user passes data to the data argument, it must be exactly the data that was used to fit the model object. Pass datasets to newdata to augment data that was not used during model fitting. This still requires that at least all predictor variable columns used to fit the model are present. If the original outcome variable used to fit the model is not included in newdata, then no .resid column will be included in the output.

Augment will often behave differently depending on whether data or newdata is given. This is because there is often information associated with training observations (such as influences or related) measures that is not meaningfully defined for new observations.

For convenience, many augment methods provide default data arguments, so that augment(fit) will return the augmented training data. In these cases, augment tries to reconstruct the original data based on the model object with varying degrees of success.

The augmented dataset is always returned as a tibble::tibble with the same number of rows as the passed dataset. This means that the passed data must be coercible to a tibble. If a predictor enters the model as part of a matrix of covariates, such as when the model formula uses splines::ns(), stats::poly(), or survival::Surv(), it is represented as a matrix column.

We are in the process of defining behaviors for models fit with various na.action arguments, but make no guarantees about behavior when data is missing at this time.

Usage

## S3 method for class 'glmRob'
augment(x, ...)

Arguments

Description

Augment accepts a model object and a dataset and adds information about each observation in the dataset. Most commonly, this includes predicted values in the .fitted column, residuals in the .resid column, and standard errors for the fitted values in a .se.fit column. New columns always begin with a . prefix to avoid overwriting columns in the original dataset.

Users may pass data to augment via either the data argument or the newdata argument. If the user passes data to the data argument, it must be exactly the data that was used to fit the model object. Pass datasets to newdata to augment data that was not used during model fitting. This still requires that at least all predictor variable columns used to fit the model are present. If the original outcome variable used to fit the model is not included in newdata, then no .resid column will be included in the output.

Augment will often behave differently depending on whether data or newdata is given. This is because there is often information associated with training observations (such as influences or related) measures that is not meaningfully defined for new observations.

For convenience, many augment methods provide default data arguments, so that augment(fit) will return the augmented training data. In these cases, augment tries to reconstruct the original data based on the model object with varying degrees of success.

The augmented dataset is always returned as a tibble::tibble with the same number of rows as the passed dataset. This means that the passed data must be coercible to a tibble. If a predictor enters the model as part of a matrix of covariates, such as when the model formula uses splines::ns(), stats::poly(), or survival::Surv(), it is represented as a matrix column.

We are in the process of defining behaviors for models fit with various na.action arguments, but make no guarantees about behavior when data is missing at this time.

Usage

## S3 method for class 'htest'
augment(x, ...)

Arguments

Details

See stats::chisq.test() for more details on how residuals are computed.

Value

A tibble::tibble() with exactly one row and columns:

See Also

augment(), stats::chisq.test()

Other htest tidiers: tidy.htest(), tidy.pairwise.htest(), tidy.power.htest()

Examples

tt <- t.test(rnorm(10))

tidy(tt)

# the glance output will be the same for each of the below tests
glance(tt)

tt <- t.test(mpg ~ am, data = mtcars)

tidy(tt)

wt <- wilcox.test(mpg ~ am, data = mtcars, conf.int = TRUE, exact = FALSE)

tidy(wt)

ct <- cor.test(mtcars$wt, mtcars$mpg)

tidy(ct)

chit <- chisq.test(xtabs(Freq ~ Sex + Class, data = as.data.frame(Titanic)))

tidy(chit)
augment(chit)

Description

Augment accepts a model object and a dataset and adds information about each observation in the dataset. Most commonly, this includes predicted values in the .fitted column, residuals in the .resid column, and standard errors for the fitted values in a .se.fit column. New columns always begin with a . prefix to avoid overwriting columns in the original dataset.

Users may pass data to augment via either the data argument or the newdata argument. If the user passes data to the data argument, it must be exactly the data that was used to fit the model object. Pass datasets to newdata to augment data that was not used during model fitting. This still requires that at least all predictor variable columns used to fit the model are present. If the original outcome variable used to fit the model is not included in newdata, then no .resid column will be included in the output.

Augment will often behave differently depending on whether data or newdata is given. This is because there is often information associated with training observations (such as influences or related) measures that is not meaningfully defined for new observations.

For convenience, many augment methods provide default data arguments, so that augment(fit) will return the augmented training data. In these cases, augment tries to reconstruct the original data based on the model object with varying degrees of success.

The augmented dataset is always returned as a tibble::tibble with the same number of rows as the passed dataset. This means that the passed data must be coercible to a tibble. If a predictor enters the model as part of a matrix of covariates, such as when the model formula uses splines::ns(), stats::poly(), or survival::Surv(), it is represented as a matrix column.

We are in the process of defining behaviors for models fit with various na.action arguments, but make no guarantees about behavior when data is missing at this time.

Usage

## S3 method for class 'ivreg'
augment(x, data = model.frame(x), newdata = NULL, ...)

Arguments

Details

This tidier currently only supports ivreg-classed objects outputted by the AER package. The ivreg package also outputs objects of class ivreg, and will be supported in a later release.

Value

A tibble::tibble() with columns:

See Also

augment(), AER::ivreg()

Other ivreg tidiers: glance.ivreg(), tidy.ivreg()

Examples

# load libraries for models and data
library(AER)

# load data
data("CigarettesSW", package = "AER")

# fit model
ivr <- ivreg(
  log(packs) ~ income | population,
  data = CigarettesSW,
  subset = year == "1995"
)

# summarize model fit with tidiers
tidy(ivr)
tidy(ivr, conf.int = TRUE)
tidy(ivr, conf.int = TRUE, instruments = TRUE)

augment(ivr)
augment(ivr, data = CigarettesSW)
augment(ivr, newdata = CigarettesSW)

glance(ivr)

Description

Augment accepts a model object and a dataset and adds information about each observation in the dataset. Most commonly, this includes predicted values in the .fitted column, residuals in the .resid column, and standard errors for the fitted values in a .se.fit column. New columns always begin with a . prefix to avoid overwriting columns in the original dataset.

Users may pass data to augment via either the data argument or the newdata argument. If the user passes data to the data argument, it must be exactly the data that was used to fit the model object. Pass datasets to newdata to augment data that was not used during model fitting. This still requires that at least all predictor variable columns used to fit the model are present. If the original outcome variable used to fit the model is not included in newdata, then no .resid column will be included in the output.

Augment will often behave differently depending on whether data or newdata is given. This is because there is often information associated with training observations (such as influences or related) measures that is not meaningfully defined for new observations.

For convenience, many augment methods provide default data arguments, so that augment(fit) will return the augmented training data. In these cases, augment tries to reconstruct the original data based on the model object with varying degrees of success.

The augmented dataset is always returned as a tibble::tibble with the same number of rows as the passed dataset. This means that the passed data must be coercible to a tibble. If a predictor enters the model as part of a matrix of covariates, such as when the model formula uses splines::ns(), stats::poly(), or survival::Surv(), it is represented as a matrix column.

We are in the process of defining behaviors for models fit with various na.action arguments, but make no guarantees about behavior when data is missing at this time.

Usage

## S3 method for class 'kmeans'
augment(x, data, ...)

Arguments

Value

A tibble::tibble() with columns:

See Also

augment(), stats::kmeans()

Other kmeans tidiers: glance.kmeans(), tidy.kmeans()

Examples

library(cluster)
library(modeldata)
library(dplyr)

data(hpc_data)

x <- hpc_data[, 2:5]

fit <- pam(x, k = 4)

tidy(fit)
glance(fit)
augment(fit, x)

Description

Augment accepts a model object and a dataset and adds information about each observation in the dataset. Most commonly, this includes predicted values in the .fitted column, residuals in the .resid column, and standard errors for the fitted values in a .se.fit column. New columns always begin with a . prefix to avoid overwriting columns in the original dataset.

Users may pass data to augment via either the data argument or the newdata argument. If the user passes data to the data argument, it must be exactly the data that was used to fit the model object. Pass datasets to newdata to augment data that was not used during model fitting. This still requires that at least all predictor variable columns used to fit the model are present. If the original outcome variable used to fit the model is not included in newdata, then no .resid column will be included in the output.

Augment will often behave differently depending on whether data or newdata is given. This is because there is often information associated with training observations (such as influences or related) measures that is not meaningfully defined for new observations.

For convenience, many augment methods provide default data arguments, so that augment(fit) will return the augmented training data. In these cases, augment tries to reconstruct the original data based on the model object with varying degrees of success.

The augmented dataset is always returned as a tibble::tibble with the same number of rows as the passed dataset. This means that the passed data must be coercible to a tibble. If a predictor enters the model as part of a matrix of covariates, such as when the model formula uses splines::ns(), stats::poly(), or survival::Surv(), it is represented as a matrix column.

We are in the process of defining behaviors for models fit with various na.action arguments, but make no guarantees about behavior when data is missing at this time.

Usage

## S3 method for class 'lm'
augment(
  x,
  data = model.frame(x),
  newdata = NULL,
  se_fit = FALSE,
  interval = c("none", "confidence", "prediction"),
  conf.level = 0.95,
  ...
)

Arguments

Details

When the modeling was performed with na.action = "na.omit" (as is the typical default), rows with NA in the initial data are omitted entirely from the augmented data frame. When the modeling was performed with na.action = "na.exclude", one should provide the original data as a second argument, at which point the augmented data will contain those rows (typically with NAs in place of the new columns). If the original data is not provided to augment() and na.action = "na.exclude", a warning is raised and the incomplete rows are dropped.

Some unusual lm objects, such as rlm from MASS, may omit .cooksd and .std.resid. gam from mgcv omits .sigma.

When newdata is supplied, only returns .fitted, .resid and .se.fit columns.

Value

A tibble::tibble() with columns:

See Also

stats::na.action

augment(), stats::predict.lm()

Other lm tidiers: augment.glm(), glance.glm(), glance.lm(), glance.summary.lm(), glance.svyglm(), tidy.glm(), tidy.lm(), tidy.lm.beta(), tidy.mlm(), tidy.summary.lm()

Examples

library(ggplot2)
library(dplyr)

mod <- lm(mpg ~ wt + qsec, data = mtcars)

tidy(mod)
glance(mod)

# coefficient plot
d <- tidy(mod, conf.int = TRUE)

ggplot(d, aes(estimate, term, xmin = conf.low, xmax = conf.high, height = 0)) +
  geom_point() +
  geom_vline(xintercept = 0, lty = 4) +
  geom_errorbarh()

# aside: There are tidy() and glance() methods for lm.summary objects too.
# this can be useful when you want to conserve memory by converting large lm
# objects into their leaner summary.lm equivalents.
s <- summary(mod)
tidy(s, conf.int = TRUE)
glance(s)

augment(mod)
augment(mod, mtcars, interval = "confidence")

# predict on new data
newdata <- mtcars %>%
  head(6) %>%
  mutate(wt = wt + 1)
augment(mod, newdata = newdata)

# ggplot2 example where we also construct 95% prediction interval

# simpler bivariate model since we're plotting in 2D
mod2 <- lm(mpg ~ wt, data = mtcars)

au <- augment(mod2, newdata = newdata, interval = "prediction")

ggplot(au, aes(wt, mpg)) +
  geom_point() +
  geom_line(aes(y = .fitted)) +
  geom_ribbon(aes(ymin = .lower, ymax = .upper), col = NA, alpha = 0.3)

# predict on new data without outcome variable. Output does not include .resid
newdata <- newdata %>%
  select(-mpg)

augment(mod, newdata = newdata)

au <- augment(mod, data = mtcars)

ggplot(au, aes(.hat, .std.resid)) +
  geom_vline(size = 2, colour = "white", xintercept = 0) +
  geom_hline(size = 2, colour = "white", yintercept = 0) +
  geom_point() +
  geom_smooth(se = FALSE)

plot(mod, which = 6)

ggplot(au, aes(.hat, .cooksd)) +
  geom_vline(xintercept = 0, colour = NA) +
  geom_abline(slope = seq(0, 3, by = 0.5), colour = "white") +
  geom_smooth(se = FALSE) +
  geom_point()

# column-wise models
a <- matrix(rnorm(20), nrow = 10)
b <- a + rnorm(length(a))
result <- lm(b ~ a)

tidy(result)

Description

Augment accepts a model object and a dataset and adds information about each observation in the dataset. Most commonly, this includes predicted values in the .fitted column, residuals in the .resid column, and standard errors for the fitted values in a .se.fit column. New columns always begin with a . prefix to avoid overwriting columns in the original dataset.

Users may pass data to augment via either the data argument or the newdata argument. If the user passes data to the data argument, it must be exactly the data that was used to fit the model object. Pass datasets to newdata to augment data that was not used during model fitting. This still requires that at least all predictor variable columns used to fit the model are present. If the original outcome variable used to fit the model is not included in newdata, then no .resid column will be included in the output.

Augment will often behave differently depending on whether data or newdata is given. This is because there is often information associated with training observations (such as influences or related) measures that is not meaningfully defined for new observations.

For convenience, many augment methods provide default data arguments, so that augment(fit) will return the augmented training data. In these cases, augment tries to reconstruct the original data based on the model object with varying degrees of success.

The augmented dataset is always returned as a tibble::tibble with the same number of rows as the passed dataset. This means that the passed data must be coercible to a tibble. If a predictor enters the model as part of a matrix of covariates, such as when the model formula uses splines::ns(), stats::poly(), or survival::Surv(), it is represented as a matrix column.

We are in the process of defining behaviors for models fit with various na.action arguments, but make no guarantees about behavior when data is missing at this time.

Usage

## S3 method for class 'lmrob'
augment(x, data = model.frame(x), newdata = NULL, se_fit = FALSE, ...)

Arguments

Details

For tidiers for robust models from the MASS package see tidy.rlm().

Value

A tibble::tibble() with columns:

See Also

robustbase::lmrob()

Other robustbase tidiers: augment.glmrob(), glance.lmrob(), tidy.glmrob(), tidy.lmrob()

Examples

if (requireNamespace("robustbase", quietly = TRUE)) {
  # load libraries for models and data
  library(robustbase)

  data(coleman)
  set.seed(0)

  m <- lmrob(Y ~ ., data = coleman)
  tidy(m)
  augment(m)
  glance(m)

  data(carrots)

  Rfit <- glmrob(cbind(success, total - success) ~ logdose + block,
    family = binomial, data = carrots, method = "Mqle",
    control = glmrobMqle.control(tcc = 1.2)
  )

  tidy(Rfit)
  augment(Rfit)
}

Description

Augment accepts a model object and a dataset and adds information about each observation in the dataset. Most commonly, this includes predicted values in the .fitted column, residuals in the .resid column, and standard errors for the fitted values in a .se.fit column. New columns always begin with a . prefix to avoid overwriting columns in the original dataset.

Users may pass data to augment via either the data argument or the newdata argument. If the user passes data to the data argument, it must be exactly the data that was used to fit the model object. Pass datasets to newdata to augment data that was not used during model fitting. This still requires that at least all predictor variable columns used to fit the model are present. If the original outcome variable used to fit the model is not included in newdata, then no .resid column will be included in the output.

Augment will often behave differently depending on whether data or newdata is given. This is because there is often information associated with training observations (such as influences or related) measures that is not meaningfully defined for new observations.

For convenience, many augment methods provide default data arguments, so that augment(fit) will return the augmented training data. In these cases, augment tries to reconstruct the original data based on the model object with varying degrees of success.

The augmented dataset is always returned as a tibble::tibble with the same number of rows as the passed dataset. This means that the passed data must be coercible to a tibble. If a predictor enters the model as part of a matrix of covariates, such as when the model formula uses splines::ns(), stats::poly(), or survival::Surv(), it is represented as a matrix column.

We are in the process of defining behaviors for models fit with various na.action arguments, but make no guarantees about behavior when data is missing at this time.

Usage

## S3 method for class 'lmRob'
augment(x, data = model.frame(x), newdata = NULL, ...)

Arguments

Details

For tidiers for robust models from the MASS package see tidy.rlm().

See Also

robust::lmRob()

Other robust tidiers: glance.glmRob(), glance.lmRob(), tidy.glmRob(), tidy.lmRob()

Examples

# load modeling library
library(robust)

# fit model
m <- lmRob(mpg ~ wt, data = mtcars)

# summarize model fit with tidiers
tidy(m)
augment(m)
glance(m)

Description

Tidy summarizes information about the components of a model. A model component might be a single term in a regression, a single hypothesis, a cluster, or a class. Exactly what tidy considers to be a model component varies across models but is usually self-evident. If a model has several distinct types of components, you will need to specify which components to return.

Usage

## S3 method for class 'loess'
augment(x, data = model.frame(x), newdata = NULL, se_fit = FALSE, ...)

Arguments

Details

When the modeling was performed with na.action = "na.omit" (as is the typical default), rows with NA in the initial data are omitted entirely from the augmented data frame. When the modeling was performed with na.action = "na.exclude", one should provide the original data as a second argument, at which point the augmented data will contain those rows (typically with NAs in place of the new columns). If the original data is not provided to augment() and na.action = "na.exclude", a warning is raised and the incomplete rows are dropped.

Note that loess objects by default will not predict on data outside of a bounding hypercube defined by the training data unless the original loess object was fit with ⁠control = loess.control(surface = \"direct\"))⁠. See stats::predict.loess() for details.

Value

A tibble::tibble() with columns:

See Also

stats::na.action

augment(), stats::loess(), stats::predict.loess()

Examples

lo <- loess(
  mpg ~ hp + wt,
  mtcars,
  control = loess.control(surface = "direct")
)

augment(lo)

# with all columns of original data
augment(lo, mtcars)

# with a new dataset
augment(lo, newdata = head(mtcars))

Description

Augment accepts a model object and a dataset and adds information about each observation in the dataset. Most commonly, this includes predicted values in the .fitted column, residuals in the .resid column, and standard errors for the fitted values in a .se.fit column. New columns always begin with a . prefix to avoid overwriting columns in the original dataset.

Users may pass data to augment via either the data argument or the newdata argument. If the user passes data to the data argument, it must be exactly the data that was used to fit the model object. Pass datasets to newdata to augment data that was not used during model fitting. This still requires that at least all predictor variable columns used to fit the model are present. If the original outcome variable used to fit the model is not included in newdata, then no .resid column will be included in the output.

Augment will often behave differently depending on whether data or newdata is given. This is because there is often information associated with training observations (such as influences or related) measures that is not meaningfully defined for new observations.

For convenience, many augment methods provide default data arguments, so that augment(fit) will return the augmented training data. In these cases, augment tries to reconstruct the original data based on the model object with varying degrees of success.

The augmented dataset is always returned as a tibble::tibble with the same number of rows as the passed dataset. This means that the passed data must be coercible to a tibble. If a predictor enters the model as part of a matrix of covariates, such as when the model formula uses splines::ns(), stats::poly(), or survival::Surv(), it is represented as a matrix column.

We are in the process of defining behaviors for models fit with various na.action arguments, but make no guarantees about behavior when data is missing at this time.

Usage

## S3 method for class 'Mclust'
augment(x, data = NULL, ...)

Arguments

Value

A tibble::tibble() with columns:

See Also

augment(), mclust::Mclust()

Other mclust tidiers: tidy.Mclust()

Examples

# load library for models and data
library(mclust)

# load data manipulation libraries
library(dplyr)
library(tibble)
library(purrr)
library(tidyr)

set.seed(27)

centers <- tibble(
  cluster = factor(1:3),
  # number points in each cluster
  num_points = c(100, 150, 50),
  # x1 coordinate of cluster center
  x1 = c(5, 0, -3),
  # x2 coordinate of cluster center
  x2 = c(-1, 1, -2)
)

points <- centers %>%
  mutate(
    x1 = map2(num_points, x1, rnorm),
    x2 = map2(num_points, x2, rnorm)
  ) %>%
  select(-num_points, -cluster) %>%
  unnest(c(x1, x2))

# fit model
m <- Mclust(points)

# summarize model fit with tidiers
tidy(m)
augment(m, points)
glance(m)

Description

Augment accepts a model object and a dataset and adds information about each observation in the dataset. Most commonly, this includes predicted values in the .fitted column, residuals in the .resid column, and standard errors for the fitted values in a .se.fit column. New columns always begin with a . prefix to avoid overwriting columns in the original dataset.

Users may pass data to augment via either the data argument or the newdata argument. If the user passes data to the data argument, it must be exactly the data that was used to fit the model object. Pass datasets to newdata to augment data that was not used during model fitting. This still requires that at least all predictor variable columns used to fit the model are present. If the original outcome variable used to fit the model is not included in newdata, then no .resid column will be included in the output.

Augment will often behave differently depending on whether data or newdata is given. This is because there is often information associated with training observations (such as influences or related) measures that is not meaningfully defined for new observations.

For convenience, many augment methods provide default data arguments, so that augment(fit) will return the augmented training data. In these cases, augment tries to reconstruct the original data based on the model object with varying degrees of success.

The augmented dataset is always returned as a tibble::tibble with the same number of rows as the passed dataset. This means that the passed data must be coercible to a tibble. If a predictor enters the model as part of a matrix of covariates, such as when the model formula uses splines::ns(), stats::poly(), or survival::Surv(), it is represented as a matrix column.

We are in the process of defining behaviors for models fit with various na.action arguments, but make no guarantees about behavior when data is missing at this time.

Usage

## S3 method for class 'mfx'
augment(
  x,
  data = model.frame(x$fit),
  newdata = NULL,
  type.predict = c("link", "response", "terms"),
  type.residuals = c("deviance", "pearson"),
  se_fit = FALSE,
  ...
)

## S3 method for class 'logitmfx'
augment(
  x,
  data = model.frame(x$fit),
  newdata = NULL,
  type.predict = c("link", "response", "terms"),
  type.residuals = c("deviance", "pearson"),
  se_fit = FALSE,
  ...
)

## S3 method for class 'negbinmfx'
augment(
  x,
  data = model.frame(x$fit),
  newdata = NULL,
  type.predict = c("link", "response", "terms"),
  type.residuals = c("deviance", "pearson"),
  se_fit = FALSE,
  ...
)

## S3 method for class 'poissonmfx'
augment(
  x,
  data = model.frame(x$fit),
  newdata = NULL,
  type.predict = c("link", "response", "terms"),
  type.residuals = c("deviance", "pearson"),
  se_fit = FALSE,
  ...
)

## S3 method for class 'probitmfx'
augment(
  x,
  data = model.frame(x$fit),
  newdata = NULL,
  type.predict = c("link", "response", "terms"),
  type.residuals = c("deviance", "pearson"),
  se_fit = FALSE,
  ...
)

Arguments

Details

This generic augment method wraps augment.glm() for applicable objects from the mfx package.

Value

A tibble::tibble() with columns:

See Also

augment.glm(), mfx::logitmfx(), mfx::negbinmfx(), mfx::poissonmfx(), mfx::probitmfx()

Other mfx tidiers: augment.betamfx(), glance.betamfx(), glance.mfx(), tidy.betamfx(), tidy.mfx()

Examples

# load libraries for models and data
library(mfx)

# get the marginal effects from a logit regression
mod_logmfx <- logitmfx(am ~ cyl + hp + wt, atmean = TRUE, data = mtcars)

tidy(mod_logmfx, conf.int = TRUE)

# compare with the naive model coefficients of the same logit call
tidy(
  glm(am ~ cyl + hp + wt, family = binomial, data = mtcars),
  conf.int = TRUE
)

augment(mod_logmfx)
glance(mod_logmfx)

# another example, this time using probit regression
mod_probmfx <- probitmfx(am ~ cyl + hp + wt, atmean = TRUE, data = mtcars)

tidy(mod_probmfx, conf.int = TRUE)
augment(mod_probmfx)
glance(mod_probmfx)

Description

Augment accepts a model object and a dataset and adds information about each observation in the dataset. Most commonly, this includes predicted values in the .fitted column, residuals in the .resid column, and standard errors for the fitted values in a .se.fit column. New columns always begin with a . prefix to avoid overwriting columns in the original dataset.

Users may pass data to augment via either the data argument or the newdata argument. If the user passes data to the data argument, it must be exactly the data that was used to fit the model object. Pass datasets to newdata to augment data that was not used during model fitting. This still requires that at least all predictor variable columns used to fit the model are present. If the original outcome variable used to fit the model is not included in newdata, then no .resid column will be included in the output.

Augment will often behave differently depending on whether data or newdata is given. This is because there is often information associated with training observations (such as influences or related) measures that is not meaningfully defined for new observations.

For convenience, many augment methods provide default data arguments, so that augment(fit) will return the augmented training data. In these cases, augment tries to reconstruct the original data based on the model object with varying degrees of success.

The augmented dataset is always returned as a tibble::tibble with the same number of rows as the passed dataset. This means that the passed data must be coercible to a tibble. If a predictor enters the model as part of a matrix of covariates, such as when the model formula uses splines::ns(), stats::poly(), or survival::Surv(), it is represented as a matrix column.

We are in the process of defining behaviors for models fit with various na.action arguments, but make no guarantees about behavior when data is missing at this time.

Usage

## S3 method for class 'mjoint'
augment(x, data = x$data, ...)

Arguments

Details

See joineRML::fitted.mjoint() and joineRML::residuals.mjoint() for more information on the difference between population-level and individual-level fitted values and residuals.

If fitting a joint model with a single longitudinal process, make sure you are using a named list to define the formula for the fixed and random effects of the longitudinal submodel.

Value

A tibble::tibble() with one row for each original observation with addition columns:

Examples

# broom only skips running these examples because the example models take a
# while to generate—they should run just fine, though!
## Not run: 


# load libraries for models and data
library(joineRML)

# fit a joint model with bivariate longitudinal outcomes
data(heart.valve)

hvd <- heart.valve[!is.na(heart.valve$log.grad) &
  !is.na(heart.valve$log.lvmi) &
  heart.valve$num <= 50, ]

fit <- mjoint(
  formLongFixed = list(
    "grad" = log.grad ~ time + sex + hs,
    "lvmi" = log.lvmi ~ time + sex
  ),
  formLongRandom = list(
    "grad" = ~ 1 | num,
    "lvmi" = ~ time | num
  ),
  formSurv = Surv(fuyrs, status) ~ age,
  data = hvd,
  inits = list("gamma" = c(0.11, 1.51, 0.80)),
  timeVar = "time"
)

# extract the survival fixed effects
tidy(fit)

# extract the longitudinal fixed effects
tidy(fit, component = "longitudinal")

# extract the survival fixed effects with confidence intervals
tidy(fit, ci = TRUE)

# extract the survival fixed effects with confidence intervals based
# on bootstrapped standard errors
bSE <- bootSE(fit, nboot = 5, safe.boot = TRUE)
tidy(fit, boot_se = bSE, ci = TRUE)

# augment original data with fitted longitudinal values and residuals
hvd2 <- augment(fit)

# extract model statistics
glance(fit)

## End(Not run)

Description

Augment accepts a model object and a dataset and adds information about each observation in the dataset. Most commonly, this includes predicted values in the .fitted column, residuals in the .resid column, and standard errors for the fitted values in a .se.fit column. New columns always begin with a . prefix to avoid overwriting columns in the original dataset.

Users may pass data to augment via either the data argument or the newdata argument. If the user passes data to the data argument, it must be exactly the data that was used to fit the model object. Pass datasets to newdata to augment data that was not used during model fitting. This still requires that at least all predictor variable columns used to fit the model are present. If the original outcome variable used to fit the model is not included in newdata, then no .resid column will be included in the output.

Augment will often behave differently depending on whether data or newdata is given. This is because there is often information associated with training observations (such as influences or related) measures that is not meaningfully defined for new observations.

For convenience, many augment methods provide default data arguments, so that augment(fit) will return the augmented training data. In these cases, augment tries to reconstruct the original data based on the model object with varying degrees of success.

The augmented dataset is always returned as a tibble::tibble with the same number of rows as the passed dataset. This means that the passed data must be coercible to a tibble. If a predictor enters the model as part of a matrix of covariates, such as when the model formula uses splines::ns(), stats::poly(), or survival::Surv(), it is represented as a matrix column.

We are in the process of defining behaviors for models fit with various na.action arguments, but make no guarantees about behavior when data is missing at this time.

Usage

## S3 method for class 'mlogit'
augment(x, data = x$model, ...)

Arguments

Details

At the moment this only works on the estimation dataset. Need to set it up to predict on another dataset.

Value

A tibble::tibble() with columns:

See Also

augment()

Other mlogit tidiers: glance.mlogit(), tidy.mlogit()

Examples

# load libraries for models and data
library(mlogit)

data("Fishing", package = "mlogit")
Fish <- dfidx(Fishing, varying = 2:9, shape = "wide", choice = "mode")

# fit model
m <- mlogit(mode ~ price + catch | income, data = Fish)

# summarize model fit with tidiers
tidy(m)
augment(m)
glance(m)

Description

Tidy summarizes information about the components of a model. A model component might be a single term in a regression, a single hypothesis, a cluster, or a class. Exactly what tidy considers to be a model component varies across models but is usually self-evident. If a model has several distinct types of components, you will need to specify which components to return.

Usage

## S3 method for class 'nlrq'
augment(x, data = NULL, newdata = NULL, ...)

Arguments

See Also

augment(), quantreg::nlrq()

Other quantreg tidiers: augment.rq(), augment.rqs(), glance.nlrq(), glance.rq(), tidy.nlrq(), tidy.rq(), tidy.rqs()

Examples

# fit model
n <- nls(mpg ~ k * e^wt, data = mtcars, start = list(k = 1, e = 2))

# summarize model fit with tidiers + visualization
tidy(n)
augment(n)
glance(n)

library(ggplot2)

ggplot(augment(n), aes(wt, mpg)) +
  geom_point() +
  geom_line(aes(y = .fitted))

newdata <- head(mtcars)
newdata$wt <- newdata$wt + 1

augment(n, newdata = newdata)

Description

Augment accepts a model object and a dataset and adds information about each observation in the dataset. Most commonly, this includes predicted values in the .fitted column, residuals in the .resid column, and standard errors for the fitted values in a .se.fit column. New columns always begin with a . prefix to avoid overwriting columns in the original dataset.

Users may pass data to augment via either the data argument or the newdata argument. If the user passes data to the data argument, it must be exactly the data that was used to fit the model object. Pass datasets to newdata to augment data that was not used during model fitting. This still requires that at least all predictor variable columns used to fit the model are present. If the original outcome variable used to fit the model is not included in newdata, then no .resid column will be included in the output.

Augment will often behave differently depending on whether data or newdata is given. This is because there is often information associated with training observations (such as influences or related) measures that is not meaningfully defined for new observations.

For convenience, many augment methods provide default data arguments, so that augment(fit) will return the augmented training data. In these cases, augment tries to reconstruct the original data based on the model object with varying degrees of success.

The augmented dataset is always returned as a tibble::tibble with the same number of rows as the passed dataset. This means that the passed data must be coercible to a tibble. If a predictor enters the model as part of a matrix of covariates, such as when the model formula uses splines::ns(), stats::poly(), or survival::Surv(), it is represented as a matrix column.

We are in the process of defining behaviors for models fit with various na.action arguments, but make no guarantees about behavior when data is missing at this time.

Usage

## S3 method for class 'nls'
augment(x, data = NULL, newdata = NULL, ...)

Arguments

Details

augment.nls does not currently support confidence intervals due to a lack of support in stats::predict.nls().

Value

A tibble::tibble() with columns:

See Also

tidy, stats::nls(), stats::predict.nls()

Other nls tidiers: glance.nls(), tidy.nls()

Examples

# fit model
n <- nls(mpg ~ k * e^wt, data = mtcars, start = list(k = 1, e = 2))

# summarize model fit with tidiers + visualization
tidy(n)
augment(n)
glance(n)

library(ggplot2)

ggplot(augment(n), aes(wt, mpg)) +
  geom_point() +
  geom_line(aes(y = .fitted))

newdata <- head(mtcars)
newdata$wt <- newdata$wt + 1

augment(n, newdata = newdata)

Description

Augment accepts a model object and a dataset and adds information about each observation in the dataset. Most commonly, this includes predicted values in the .fitted column, residuals in the .resid column, and standard errors for the fitted values in a .se.fit column. New columns always begin with a . prefix to avoid overwriting columns in the original dataset.

Users may pass data to augment via either the data argument or the newdata argument. If the user passes data to the data argument, it must be exactly the data that was used to fit the model object. Pass datasets to newdata to augment data that was not used during model fitting. This still requires that at least all predictor variable columns used to fit the model are present. If the original outcome variable used to fit the model is not included in newdata, then no .resid column will be included in the output.

Augment will often behave differently depending on whether data or newdata is given. This is because there is often information associated with training observations (such as influences or related) measures that is not meaningfully defined for new observations.

For convenience, many augment methods provide default data arguments, so that augment(fit) will return the augmented training data. In these cases, augment tries to reconstruct the original data based on the model object with varying degrees of success.

The augmented dataset is always returned as a tibble::tibble with the same number of rows as the passed dataset. This means that the passed data must be coercible to a tibble. If a predictor enters the model as part of a matrix of covariates, such as when the model formula uses splines::ns(), stats::poly(), or survival::Surv(), it is represented as a matrix column.

We are in the process of defining behaviors for models fit with various na.action arguments, but make no guarantees about behavior when data is missing at this time.

Usage

## S3 method for class 'pam'
augment(x, data = NULL, ...)

Arguments

Value

A tibble::tibble() with columns:

See Also

augment(), cluster::pam()

Other pam tidiers: glance.pam(), tidy.pam()

Examples

# load libraries for models and data
library(dplyr)
library(ggplot2)
library(cluster)
library(modeldata)
data(hpc_data)

x <- hpc_data[, 2:5]
p <- pam(x, k = 4)

# summarize model fit with tidiers + visualization
tidy(p)
glance(p)
augment(p, x)

augment(p, x) %>%
  ggplot(aes(compounds, input_fields)) +
  geom_point(aes(color = .cluster)) +
  geom_text(aes(label = cluster), data = tidy(p), size = 10)

Description

Augment accepts a model object and a dataset and adds information about each observation in the dataset. Most commonly, this includes predicted values in the .fitted column, residuals in the .resid column, and standard errors for the fitted values in a .se.fit column. New columns always begin with a . prefix to avoid overwriting columns in the original dataset.

Users may pass data to augment via either the data argument or the newdata argument. If the user passes data to the data argument, it must be exactly the data that was used to fit the model object. Pass datasets to newdata to augment data that was not used during model fitting. This still requires that at least all predictor variable columns used to fit the model are present. If the original outcome variable used to fit the model is not included in newdata, then no .resid column will be included in the output.

Augment will often behave differently depending on whether data or newdata is given. This is because there is often information associated with training observations (such as influences or related) measures that is not meaningfully defined for new observations.

For convenience, many augment methods provide default data arguments, so that augment(fit) will return the augmented training data. In these cases, augment tries to reconstruct the original data based on the model object with varying degrees of success.

The augmented dataset is always returned as a tibble::tibble with the same number of rows as the passed dataset. This means that the passed data must be coercible to a tibble. If a predictor enters the model as part of a matrix of covariates, such as when the model formula uses splines::ns(), stats::poly(), or survival::Surv(), it is represented as a matrix column.

We are in the process of defining behaviors for models fit with various na.action arguments, but make no guarantees about behavior when data is missing at this time.

Usage

## S3 method for class 'plm'
augment(x, data = model.frame(x), ...)

Arguments

Value

A tibble::tibble() with columns:

See Also

augment(), plm::plm()

Other plm tidiers: glance.plm(), tidy.plm()

Examples

# load libraries for models and data
library(plm)

# load data
data("Produc", package = "plm")

# fit model
zz <- plm(log(gsp) ~ log(pcap) + log(pc) + log(emp) + unemp,
  data = Produc, index = c("state", "year")
)

# summarize model fit with tidiers
summary(zz)

tidy(zz)
tidy(zz, conf.int = TRUE)
tidy(zz, conf.int = TRUE, conf.level = 0.9)

augment(zz)
glance(zz)

Description

Augment accepts a model object and a dataset and adds information about each observation in the dataset. Most commonly, this includes predicted values in the .fitted column, residuals in the .resid column, and standard errors for the fitted values in a .se.fit column. New columns always begin with a . prefix to avoid overwriting columns in the original dataset.

Users may pass data to augment via either the data argument or the newdata argument. If the user passes data to the data argument, it must be exactly the data that was used to fit the model object. Pass datasets to newdata to augment data that was not used during model fitting. This still requires that at least all predictor variable columns used to fit the model are present. If the original outcome variable used to fit the model is not included in newdata, then no .resid column will be included in the output.

Augment will often behave differently depending on whether data or newdata is given. This is because there is often information associated with training observations (such as influences or related) measures that is not meaningfully defined for new observations.

For convenience, many augment methods provide default data arguments, so that augment(fit) will return the augmented training data. In these cases, augment tries to reconstruct the original data based on the model object with varying degrees of success.

The augmented dataset is always returned as a tibble::tibble with the same number of rows as the passed dataset. This means that the passed data must be coercible to a tibble. If a predictor enters the model as part of a matrix of covariates, such as when the model formula uses splines::ns(), stats::poly(), or survival::Surv(), it is represented as a matrix column.

We are in the process of defining behaviors for models fit with various na.action arguments, but make no guarantees about behavior when data is missing at this time.

Usage

## S3 method for class 'poLCA'
augment(x, data = NULL, ...)

Arguments

Details

If the data argument is given, those columns are included in the output (only rows for which predictions could be made). Otherwise, the y element of the poLCA object, which contains the manifest variables used to fit the model, are used, along with any covariates, if present, in x.

Note that while the probability of all the classes (not just the predicted modal class) can be found in the posterior element, these are not included in the augmented output.

Value

A tibble::tibble() with columns:

See Also

augment(), poLCA::poLCA()

Other poLCA tidiers: glance.poLCA(), tidy.poLCA()

Examples

# load libraries for models and data
library(poLCA)
library(dplyr)

# generate data
data(values)

f <- cbind(A, B, C, D) ~ 1

# fit model
M1 <- poLCA(f, values, nclass = 2, verbose = FALSE)

M1

# summarize model fit with tidiers + visualization
tidy(M1)
augment(M1)
glance(M1)

library(ggplot2)

ggplot(tidy(M1), aes(factor(class), estimate, fill = factor(outcome))) +
  geom_bar(stat = "identity", width = 1) +
  facet_wrap(~variable)

# three-class model with a single covariate.
data(election)

f2a <- cbind(
  MORALG, CARESG, KNOWG, LEADG, DISHONG, INTELG,
  MORALB, CARESB, KNOWB, LEADB, DISHONB, INTELB
) ~ PARTY

nes2a <- poLCA(f2a, election, nclass = 3, nrep = 5, verbose = FALSE)

td <- tidy(nes2a)
td

ggplot(td, aes(outcome, estimate, color = factor(class), group = class)) +
  geom_line() +
  facet_wrap(~variable, nrow = 2) +
  theme(axis.text.x = element_text(angle = 90, hjust = 1))

au <- augment(nes2a)

au

count(au, .class)

# if the original data is provided, it leads to NAs in new columns
# for rows that weren't predicted
au2 <- augment(nes2a, data = election)

au2

dim(au2)

Description

Augment accepts a model object and a dataset and adds information about each observation in the dataset. Most commonly, this includes predicted values in the .fitted column, residuals in the .resid column, and standard errors for the fitted values in a .se.fit column. New columns always begin with a . prefix to avoid overwriting columns in the original dataset.

Users may pass data to augment via either the data argument or the newdata argument. If the user passes data to the data argument, it must be exactly the data that was used to fit the model object. Pass datasets to newdata to augment data that was not used during model fitting. This still requires that at least all predictor variable columns used to fit the model are present. If the original outcome variable used to fit the model is not included in newdata, then no .resid column will be included in the output.

Augment will often behave differently depending on whether data or newdata is given. This is because there is often information associated with training observations (such as influences or related) measures that is not meaningfully defined for new observations.

For convenience, many augment methods provide default data arguments, so that augment(fit) will return the augmented training data. In these cases, augment tries to reconstruct the original data based on the model object with varying degrees of success.

The augmented dataset is always returned as a tibble::tibble with the same number of rows as the passed dataset. This means that the passed data must be coercible to a tibble. If a predictor enters the model as part of a matrix of covariates, such as when the model formula uses splines::ns(), stats::poly(), or survival::Surv(), it is represented as a matrix column.

We are in the process of defining behaviors for models fit with various na.action arguments, but make no guarantees about behavior when data is missing at this time.

Usage

## S3 method for class 'polr'
augment(
  x,
  data = model.frame(x),
  newdata = NULL,
  type.predict = c("class"),
  ...
)

Arguments

See Also

tidy(), MASS::polr()

Other ordinal tidiers: augment.clm(), glance.clm(), glance.clmm(), glance.polr(), glance.svyolr(), tidy.clm(), tidy.clmm(), tidy.polr(), tidy.svyolr()

Examples

# load libraries for models and data
library(MASS)

# fit model
fit <- polr(Sat ~ Infl + Type + Cont, weights = Freq, data = housing)

# summarize model fit with tidiers
tidy(fit, exponentiate = TRUE, conf.int = TRUE)

glance(fit)
augment(fit, type.predict = "class")

fit2 <- polr(factor(gear) ~ am + mpg + qsec, data = mtcars)

tidy(fit, p.values = TRUE)

Description

Augment accepts a model object and a dataset and adds information about each observation in the dataset. Most commonly, this includes predicted values in the .fitted column, residuals in the .resid column, and standard errors for the fitted values in a .se.fit column. New columns always begin with a . prefix to avoid overwriting columns in the original dataset.

Users may pass data to augment via either the data argument or the newdata argument. If the user passes data to the data argument, it must be exactly the data that was used to fit the model object. Pass datasets to newdata to augment data that was not used during model fitting. This still requires that at least all predictor variable columns used to fit the model are present. If the original outcome variable used to fit the model is not included in newdata, then no .resid column will be included in the output.

Augment will often behave differently depending on whether data or newdata is given. This is because there is often information associated with training observations (such as influences or related) measures that is not meaningfully defined for new observations.

For convenience, many augment methods provide default data arguments, so that augment(fit) will return the augmented training data. In these cases, augment tries to reconstruct the original data based on the model object with varying degrees of success.

The augmented dataset is always returned as a tibble::tibble with the same number of rows as the passed dataset. This means that the passed data must be coercible to a tibble. If a predictor enters the model as part of a matrix of covariates, such as when the model formula uses splines::ns(), stats::poly(), or survival::Surv(), it is represented as a matrix column.

We are in the process of defining behaviors for models fit with various na.action arguments, but make no guarantees about behavior when data is missing at this time.

Usage

## S3 method for class 'prcomp'
augment(x, data = NULL, newdata, ...)

Arguments

Value

A tibble::tibble containing the original data along with additional columns containing each observation's projection into PCA space.

See Also

stats::prcomp(), svd_tidiers

Other svd tidiers: tidy.prcomp(), tidy_irlba(), tidy_svd()

Description

Augment accepts a model object and a dataset and adds information about each observation in the dataset. Most commonly, this includes predicted values in the .fitted column, residuals in the .resid column, and standard errors for the fitted values in a .se.fit column. New columns always begin with a . prefix to avoid overwriting columns in the original dataset.

Users may pass data to augment via either the data argument or the newdata argument. If the user passes data to the data argument, it must be exactly the data that was used to fit the model object. Pass datasets to newdata to augment data that was not used during model fitting. This still requires that at least all predictor variable columns used to fit the model are present. If the original outcome variable used to fit the model is not included in newdata, then no .resid column will be included in the output.

Augment will often behave differently depending on whether data or newdata is given. This is because there is often information associated with training observations (such as influences or related) measures that is not meaningfully defined for new observations.

For convenience, many augment methods provide default data arguments, so that augment(fit) will return the augmented training data. In these cases, augment tries to reconstruct the original data based on the model object with varying degrees of success.

The augmented dataset is always returned as a tibble::tibble with the same number of rows as the passed dataset. This means that the passed data must be coercible to a tibble. If a predictor enters the model as part of a matrix of covariates, such as when the model formula uses splines::ns(), stats::poly(), or survival::Surv(), it is represented as a matrix column.

We are in the process of defining behaviors for models fit with various na.action arguments, but make no guarantees about behavior when data is missing at this time.

Usage

## S3 method for class 'rlm'
augment(x, data = model.frame(x), newdata = NULL, se_fit = FALSE, ...)

Arguments

Value

A tibble::tibble() with columns:

See Also

MASS::rlm()

Other rlm tidiers: glance.rlm(), tidy.rlm()

Examples

# load libraries for models and data
library(MASS)

# fit model
r <- rlm(stack.loss ~ ., stackloss)

# summarize model fit with tidiers
tidy(r)
augment(r)
glance(r)

Description

Augment accepts a model object and a dataset and adds information about each observation in the dataset. Most commonly, this includes predicted values in the .fitted column, residuals in the .resid column, and standard errors for the fitted values in a .se.fit column. New columns always begin with a . prefix to avoid overwriting columns in the original dataset.

Users may pass data to augment via either the data argument or the newdata argument. If the user passes data to the data argument, it must be exactly the data that was used to fit the model object. Pass datasets to newdata to augment data that was not used during model fitting. This still requires that at least all predictor variable columns used to fit the model are present. If the original outcome variable used to fit the model is not included in newdata, then no .resid column will be included in the output.

Augment will often behave differently depending on whether data or newdata is given. This is because there is often information associated with training observations (such as influences or related) measures that is not meaningfully defined for new observations.

For convenience, many augment methods provide default data arguments, so that augment(fit) will return the augmented training data. In these cases, augment tries to reconstruct the original data based on the model object with varying degrees of success.

The augmented dataset is always returned as a tibble::tibble with the same number of rows as the passed dataset. This means that the passed data must be coercible to a tibble. If a predictor enters the model as part of a matrix of covariates, such as when the model formula uses splines::ns(), stats::poly(), or survival::Surv(), it is represented as a matrix column.

We are in the process of defining behaviors for models fit with various na.action arguments, but make no guarantees about behavior when data is missing at this time.

Usage

## S3 method for class 'rma'
augment(x, interval = c("prediction", "confidence"), ...)

Arguments

Value

A tibble::tibble() with columns:

Examples

# load modeling library
library(metafor)

# generate data and fit
df <-
  escalc(
    measure = "RR",
    ai = tpos,
    bi = tneg,
    ci = cpos,
    di = cneg,
    data = dat.bcg
  )

meta_analysis <- rma(yi, vi, data = df, method = "EB")

# summarize model fit with tidiers
augment(meta_analysis)

Description

Augment accepts a model object and a dataset and adds information about each observation in the dataset. Most commonly, this includes predicted values in the .fitted column, residuals in the .resid column, and standard errors for the fitted values in a .se.fit column. New columns always begin with a . prefix to avoid overwriting columns in the original dataset.

Users may pass data to augment via either the data argument or the newdata argument. If the user passes data to the data argument, it must be exactly the data that was used to fit the model object. Pass datasets to newdata to augment data that was not used during model fitting. This still requires that at least all predictor variable columns used to fit the model are present. If the original outcome variable used to fit the model is not included in newdata, then no .resid column will be included in the output.

Augment will often behave differently depending on whether data or newdata is given. This is because there is often information associated with training observations (such as influences or related) measures that is not meaningfully defined for new observations.

For convenience, many augment methods provide default data arguments, so that augment(fit) will return the augmented training data. In these cases, augment tries to reconstruct the original data based on the model object with varying degrees of success.

The augmented dataset is always returned as a tibble::tibble with the same number of rows as the passed dataset. This means that the passed data must be coercible to a tibble. If a predictor enters the model as part of a matrix of covariates, such as when the model formula uses splines::ns(), stats::poly(), or survival::Surv(), it is represented as a matrix column.

We are in the process of defining behaviors for models fit with various na.action arguments, but make no guarantees about behavior when data is missing at this time.

Usage

## S3 method for class 'rq'
augment(x, data = model.frame(x), newdata = NULL, ...)

Arguments

Details

Depending on the arguments passed on to predict.rq via ..., a confidence interval is also calculated on the fitted values resulting in columns .lower and .upper. Does not provide confidence intervals when data is specified via the newdata argument.

Value

A tibble::tibble() with columns:

See Also

augment, quantreg::rq(), quantreg::predict.rq()

Other quantreg tidiers: augment.nlrq(), augment.rqs(), glance.nlrq(), glance.rq(), tidy.nlrq(), tidy.rq(), tidy.rqs()

Examples

# load modeling library and data
library(quantreg)

data(stackloss)

# median (l1) regression fit for the stackloss data.
mod1 <- rq(stack.loss ~ stack.x, .5)

# weighted sample median
mod2 <- rq(rnorm(50) ~ 1, weights = runif(50))

# summarize model fit with tidiers
tidy(mod1)
glance(mod1)
augment(mod1)

tidy(mod2)
glance(mod2)
augment(mod2)

# varying tau to generate an rqs object
mod3 <- rq(stack.loss ~ stack.x, tau = c(.25, .5))

tidy(mod3)
augment(mod3)

# glance cannot handle rqs objects like `mod3`--use a purrr
# `map`-based workflow instead

Description

Augment accepts a model object and a dataset and adds information about each observation in the dataset. Most commonly, this includes predicted values in the .fitted column, residuals in the .resid column, and standard errors for the fitted values in a .se.fit column. New columns always begin with a . prefix to avoid overwriting columns in the original dataset.

Users may pass data to augment via either the data argument or the newdata argument. If the user passes data to the data argument, it must be exactly the data that was used to fit the model object. Pass datasets to newdata to augment data that was not used during model fitting. This still requires that at least all predictor variable columns used to fit the model are present. If the original outcome variable used to fit the model is not included in newdata, then no .resid column will be included in the output.

Augment will often behave differently depending on whether data or newdata is given. This is because there is often information associated with training observations (such as influences or related) measures that is not meaningfully defined for new observations.

For convenience, many augment methods provide default data arguments, so that augment(fit) will return the augmented training data. In these cases, augment tries to reconstruct the original data based on the model object with varying degrees of success.