Package {bootGOF}

Usage

GOF_glm_info_extractor$yhat(model)

Usage

GOF_glm_info_extractor$y_minus_yhat(model)

Arguments

model: see GOF_model_info_extractor

Returns

Method `beta_x_covariates()`

Usage

GOF_glm_info_extractor$beta_x_covariates(model)

Arguments

model: see GOF_model_info_extractor

Returns

Method `clone()`

The objects of this class are cloneable with this method.

Usage

GOF_glm_info_extractor$clone(deep = FALSE)

Arguments

deep: Whether to make a deep clone.

Implements the "interface" GOF_model_simulator for for generalized linear models

Description

after the GLM was fitted the distribution of the of the dependent variable is fully specified and used here to generate new depenedent variables that follow model

Methods

Method `resample_y()`

Usage

GOF_glm_sim_param$resample_y(model)

Arguments

model: see GOF_model_simulator

Returns

Method `clone()`

The objects of this class are cloneable with this method.

Usage

GOF_glm_sim_param$clone(deep = FALSE)

Arguments

deep: Whether to make a deep clone.

Implements the "interface" GOF_model_trainer for for generalized linear models

Description

refits an object of class "glm" to a new data set

Methods

Public methods

GOF_glm_trainer$refit()
GOF_glm_trainer$clone()

Method `refit()`

Usage

GOF_glm_trainer$refit(model, data)

Arguments

model: see GOF_model_trainer
data: see GOF_model_trainer

Returns

Method `clone()`

The objects of this class are cloneable with this method.

Usage

GOF_glm_trainer$clone(deep = FALSE)

Arguments

deep: Whether to make a deep clone.

Implements the "interface" GOF_model_info_extractor for linear models

Description

This class is specialized in extracting various information from an object of class "lm"

Super class

bootGOF::GOF_model_info_extractor -> GOF_lm_info_extractor

Methods

Public methods

GOF_lm_info_extractor$yhat()
GOF_lm_info_extractor$y_minus_yhat()
GOF_lm_info_extractor$beta_x_covariates()
GOF_lm_info_extractor$clone()

Inherited methods

Method `yhat()`

Usage

GOF_lm_info_extractor$yhat(model)

Arguments

model: see GOF_model_info_extractor

Returns

Method `y_minus_yhat()`

Usage

GOF_lm_info_extractor$y_minus_yhat(model)

Arguments

model: see GOF_model_info_extractor

Returns

Method `beta_x_covariates()`

Usage

GOF_lm_info_extractor$beta_x_covariates(model)

Arguments

model: see GOF_model_info_extractor

Returns

Method `clone()`

The objects of this class are cloneable with this method.

Usage

GOF_lm_info_extractor$clone(deep = FALSE)

Arguments

deep: Whether to make a deep clone.

Implements the "interface" GOF_model_simulator for for linear models

Description

after the classical linear model was fitted the normal distribution of the of the dependent variable is fully specified and used here to generate new depenedent variables that follow model

Methods

Public methods

GOF_lm_sim_param$resample_y()
GOF_lm_sim_param$clone()

Method `resample_y()`

generates/resamples the dependent variables based on the parameteric nature defined by model

Usage

GOF_lm_sim_param$resample_y(model)

Arguments

model: see GOF_model_simulator

Returns

Method `clone()`

The objects of this class are cloneable with this method.

Usage

GOF_lm_sim_param$clone(deep = FALSE)

Arguments

deep: Whether to make a deep clone.

Implements the "interface" GOF_model_trainer for for linear models

Description

refits an object of class "lm" to a new data set

Methods

Public methods

GOF_lm_trainer$refit()
GOF_lm_trainer$clone()

Method `refit()`

Usage

GOF_lm_trainer$refit(model, data)

Arguments

model: see GOF_model_trainer
data: see GOF_model_trainer

Returns

Method `clone()`

The objects of this class are cloneable with this method.

Usage

GOF_lm_trainer$clone(deep = FALSE)

Arguments

deep: Whether to make a deep clone.

Convenience function for creating a GOF-test for statistical models

Description

Simplifies the creation of an instance of GOF_model_test, the actual work horse for performing a goodness-of-fit-test.

Usage

GOF_model(
  model,
  data,
  nmb_boot_samples,
  simulator_type,
  y_name,
  Rn1_statistic,
  gof_model_resample_class = GOF_model_resample,
  gof_model_test_class = GOF_model_test
)

Arguments

model

of class 'lm' or 'glm'. Caution with MASS::glm.nb, see vignette 'New-Models' for more details.

data

nmb_boot_samples

simulator_type

either "parameteric" or "semi_parameteric_rademacher"

y_name

Rn1_statistic

gof_model_resample_class

no need to change this parameter. Here the class used for resampling the model (GOF_model_resample) is injected. This parameter simply makes it easier to test the convenience function properly.

gof_model_test_class

no need to change this parameter. Here the class used for performing the GOF test (GOF_model_test) is injected. This parameter simply makes it easier to test the convenience function properly.

Value

instance of GOF_model_test

Examples

set.seed(1)
N <- 100
X1 <- rnorm(N)
X2 <- rnorm(N)
d <- data.frame(
  y = rpois(n = N, lambda = exp(4 + X1 * 2 + X2 * 6)),
  x1 = X1,
  x2 = X2)
fit <- glm(y ~ x1, data = d, family = poisson())
mt <- GOF_model(
  model = fit,
  data = d,
  nmb_boot_samples = 100,
  simulator_type = "parametric",
  y_name = "y",
  Rn1_statistic = Rn1_KS$new())
mt$get_pvalue()
fit <- glm(y ~ x1 + x2, data = d, family = poisson())
mt <- GOF_model(
  model = fit,
  data = d,
  nmb_boot_samples = 100,
  simulator_type = "parametric",
  y_name = "y",
  Rn1_statistic = Rn1_KS$new())
mt$get_pvalue()

R6 Class representing model information

Description

R6 does not offer interfaces. Hence all methods are considered as abstract.

Methods

Public methods

GOF_model_info_extractor$yhat()
GOF_model_info_extractor$y_minus_yhat()
GOF_model_info_extractor$beta_x_covariates()
GOF_model_info_extractor$clone()

Method `yhat()`

Abstract function that estimates/predicts the the dependent variable in model

Usage

GOF_model_info_extractor$yhat(model)

Arguments

model: fitted model

Returns

estimate/prediction of the dependent variable fitted by model

Method `y_minus_yhat()`

abstract function that calculates the residuals on the scale of the dependent variable.

Usage

GOF_model_info_extractor$y_minus_yhat(model)

Arguments

model: fitted model

Returns

residuals on the scale of the dependent variable

Method `beta_x_covariates()`

abstract function that calculates the inner product of estimated parameters and the independent variables.

Usage

GOF_model_info_extractor$beta_x_covariates(model)

Arguments

model: fitted model

Returns

inner product of the estimated parameters and the independent variables.

Method `clone()`

The objects of this class are cloneable with this method.

Usage

GOF_model_info_extractor$clone(deep = FALSE)

Arguments

deep: Whether to make a deep clone.

R6 Class representing the resampling scheme for Goodness-of-fit-tests for (linear) models

Description

Class is able to resample model fit, i.e. generate a new data set and refit the model to the new data.

Methods

Public methods

GOF_model_resample$new()
GOF_model_resample$resample()
GOF_model_resample$clone()

Method `new()`

Usage

GOF_model_resample$new(gof_model_simulator, gof_model_trainer)

Arguments

gof_model_simulator: an instance that implements GOF_model_simulator
gof_model_trainer: an instance that implements GOF_model_trainer

Returns

No explicit return

Method `resample()`

resamples the dependent variable in data and refits model to that new data set

Usage

GOF_model_resample$resample(model, data, y_name)

Arguments

model: fitted model based on data
data: used to fit model
y_name: string specifying the name of the dependent variable

Returns

a resampled version of model

Method `clone()`

The objects of this class are cloneable with this method.

Usage

GOF_model_resample$clone(deep = FALSE)

Arguments

deep: Whether to make a deep clone.

R6 Class representing a generator/resample of the dependent variable

Description

R6 does not offer interfaces. Hence all methods are considered as abstract.

Methods

Public methods

GOF_model_simulator$resample_y()
GOF_model_simulator$clone()

Method `resample_y()`

Abstract function that resamples/generates the dependent variable

Usage

GOF_model_simulator$resample_y(model)

Arguments

model: fitted model

Returns

generates the dependent variable according to the model

Method `clone()`

The objects of this class are cloneable with this method.

Usage

GOF_model_simulator$clone(deep = FALSE)

Arguments

deep: Whether to make a deep clone.

R6 Class representing the Goodness-of-Fit test for (linear) models.

Description

This class can test the null hypothesis that data follows a particular linear model, i.e. classical linear models, generalized linear models or models of the type m(\beta^\top X) + \epsilon.

Methods

Public methods

GOF_model_test$new()
GOF_model_test$get_Rn1_org()
GOF_model_test$get_Rn1_boot()
GOF_model_test$get_pvalue()
GOF_model_test$clone()

Method `new()`

Usage

GOF_model_test$new(
  model,
  data,
  nmb_boot_samples,
  y_name,
  Rn1_statistic,
  gof_model_info_extractor,
  gof_model_resample
)

Arguments

model: a fitted model
data: used to fit model
nmb_boot_samples: integer specifying the number of bootstrap samples to perform
y_name: string specifying the name of the dependent variable in in data
Rn1_statistic: statistic used to map the marked empirical process to the real line. Needs to be an instance of the class that implements Rn1_statistic
gof_model_info_extractor: an instance that implements GOF_model_info_extractor in order to apply it to model
gof_model_resample: an instance that implements GOF_model_resample in order to apply it to model

Returns

An instance of the Class

Method `get_Rn1_org()`

calculates the marked empricial process for model

Usage

GOF_model_test$get_Rn1_org()

Returns

vector ordered by the inner product of the estimated parameter and the independent variables

Method `get_Rn1_boot()`

calculates the marked empricial process for the resampled versions of model

Usage

GOF_model_test$get_Rn1_boot()

Returns

list of length nmb_boot_samples where every element is a vector ordered by the inner product of the estimated parameter and the dependent variables

Method `get_pvalue()`

p-value for Goodness-of-Fit-test for model

Usage

GOF_model_test$get_pvalue()

Returns

p-value for the null hypothesis that the dependent variable was generated according to model

Method `clone()`

The objects of this class are cloneable with this method.

Usage

GOF_model_test$clone(deep = FALSE)

Arguments

deep: Whether to make a deep clone.

R6 Class representing a trainer for fitting models

Description

R6 does not offer interfaces. Hence all methods are considered as abstract.

Methods

Public methods

GOF_model_trainer$refit()
GOF_model_trainer$clone()

Method `refit()`

Abstract function refits the model to a new data set

Usage

GOF_model_trainer$refit(model, data)

Arguments

model: fitted model
data: used for refitting the model

Returns

model refitted on data

Method `clone()`

The objects of this class are cloneable with this method.

Usage

GOF_model_trainer$clone(deep = FALSE)

Arguments

deep: Whether to make a deep clone.

Implements the "interface" GOF_model_simulator in a semi-parametric fashion

Description

This is a model agnostic resampling class, where Rademacher random variables are used to add or substract the residuals from the fitted values.

Methods

Public methods

GOF_sim_wild_rademacher$new()
GOF_sim_wild_rademacher$resample_y()
GOF_sim_wild_rademacher$clone()

Method `new()`

Usage

GOF_sim_wild_rademacher$new(gof_model_info_extractor)

Arguments

gof_model_info_extractor: the info extractor that is used to derive the residuals and fitted values for resampling.

Method `resample_y()`

a wild bootstrap using Rademacher random variables to resample the dependent variable

Usage

GOF_sim_wild_rademacher$resample_y(model)

Arguments

model: see GOF_model_simulator

Returns