Automated Multiple Regression Modelling

Share:

Description

Automated Multiple Regression Modelling

Usage

1
2
3
run_model(outcome, block1, ..., dataset, type = "gaussian",
  assumptions.check = T, outliers.check = "significant",
  transform.outcome = F)

Arguments

outcome

The dependent variable of the hierarchical model

block1

A character vector, with names of variables. The first block of independent variables.

...

A character vector, with names of variables. Subsequent blocks of independent variables.

dataset

A data frame containing variables refered to in formulas, passed to data argument of lm

type

Family argument to pass to glm. Specify "binomial" for binary logistic regression models.

assumptions.check

Boolean, if TRUE, then assumption checks are run and output is produced. If FALSE, only model summary and coefficient tables are produced.

outliers.check

Determines how many observations to display for outliers check. Default is significant observations. "All" shows all residual and Cook's D values.

transform.outcome

A boolean. If TRUE, a variable transformation of the outcome is substituted in the final model if outcome is non-normal. NOT IMPLEMENTED YET.

Details

Calls other functions to generate model objects and test them, given specified model parameters and other options. Formatted output is produced via model_output

Examples

1
2
run_model("y", c("lag.quarterly.revenue"), c("price.index", "income.level"),
dataset=freeny)