modelObj (version 4.2)

buildModelObj: Create an Object of Class modelObj

Description

A utility function to transfer user defined models and estimation methods to an object of class modelObj.

Usage

buildModelObj(
  model,
  solver.method = NULL,
  solver.args = NULL,
  predict.method = NULL,
  predict.args = NULL
)

Arguments

model

An object of class formula; the model.

solver.method

An object of class character specifying the name of the R function to be used to obtain parameter estimates. Or, the function to be used to obtain parameter estimates. For example, `lm', `glm', or `rpart'. The specified modeling function MUST have a corresponding predict method.

solver.args

An object of class list containing additional arguments to be sent to solver.method. Arguments must be provided as a list, where the name of each element matches a formal argument of solver.method. For example, if a logistic regression using glm is desired, $$solver.method = ``glm"$$ $$solver.args = list(``family"=binomial)$$

A solver.method can takes formal arguments 'formula' and 'data' as inputs, such as lm and glm. Some R methods do not use formal names 'formula' and 'data'; a user can indicate if a different naming convention is used for these two input arguments. For example, if a method expects the formula object to be passed through input variable x, solver.args <- list("x"="formula")

A solver.method can also take formal arguments 'x' and 'y' as inputs, such as glmnet. Some R methods do not use formal names 'x' and 'y' to indicate the covariate and response; a user can indicate if a different naming convention is used for these two input arguments. For example, if a method expects the covariate matrix to be passed through input variable X, solver.args <- list("X"="x")

predict.method

A character. The name of the R function or the function to be used to obtain predictions. For example, `predict.lm', `predict', or `predict.glm'. If no function is explicitly given, the generic predict is assumed. For many methods, the generic method is appropriate.

predict.args

A list. Additional arguments to be sent to predict.method. This must be provided as a list, where the name of each element matches a formal argument of predict.method. For example, if a logistic regression using glm was used to fit the model formula object and predictions on the scale of the response are desired, $$predict.method = ``predict.glm"$$ $$predict.args = list(``type"=``response").$$

It is assumed that the predict.method has formal arguments ``object" and ``newdata". If predict.method does not use these formal arguments, predict.args must explicitly indicate the variable names used for these inputs. For example, list(``newx"=``newdata") if the new data is passed to predict.method through input argument ``newx".

Value

An object of class modelObjFormula or modelObjXY, which inherit directly from modelObj.

Details

Unless changed by the user in solver.args and/or predict.args, default settings are assumed for the specified regression and prediction methods.

Examples

# NOT RUN {
   #----------------------------------------------------#
   # Create modeling object using a formula
   #----------------------------------------------------#
   mo <- buildModelObj(model=Y ~ X1 + X2 + X3 + X4,
                       solver.method='lm', 
                       predict.method='predict.lm',
                       predict.args=list(type='response'))
# }