flexCountReg (version 0.1.1)

regCompTable: Create a Table Comparing Regression Models with AIC, BIC, and McFadden's Pseudo-R-Squared

Description

This function creates tables comparing the flexCountReg package models supplied to the function.

Usage

regCompTable(
  models,
  coefs = TRUE,
  AIC = TRUE,
  BIC = TRUE,
  RSquare = TRUE,
  tableType = "tibble",
  digits = 3
)

Value

A table comparing the models supplied to the function in the format specified The table includes coefficients, standard errors, statistical significance, AIC, BIC, and McFadden's Pseudo-R-Squared. Table formats include: "tibble", "gt", and "latex".

Arguments

models

A named list of fitted flexCountReg model objects. This must include 2 or more models.

coefs

A logical. The default value `TRUE` indicates that the coefficients from the models should be included in the table of comparisons.

AIC

A logical. The default value `TRUE` indicates that AIC values for the models should be included.

BIC

A logical. The default value `TRUE` indicates that BIC values for the models should be included.

RSquare

A logical. The default value `TRUE` indicates that the McFadden's Pseudo-R-Squared statistic (comparing against a Poisson regression model) should be included.

tableType

The type of table format to return. Options include "tibble" for returning the table as a tibble, "gt" for a gt table object, or "latex" for a latex table. The default is "tibble".

digits

An integer value indicating the number of decimals to round the table values to.

Examples

# Comparing the NBP model with the NB2 and NB1 models
data("washington_roads")
washington_roads$AADTover10k <- ifelse(washington_roads$AADT>10000,1,0)

nb.1 <- countreg(Total_crashes ~ lnaadt + lnlength + speed50 +
                    ShouldWidth04 + AADTover10k,
                    data=washington_roads, family = 'NB1', method = 'NM')
                    
nb.2 <- countreg(Total_crashes ~ lnaadt + lnlength + speed50 +
                    ShouldWidth04 + AADTover10k,
                    data=washington_roads, family = 'NB2', method = 'NM')
                    
                    
nb.p <- countreg(Total_crashes ~ lnaadt + lnlength + speed50 +
                    ShouldWidth04 + AADTover10k,
                    data=washington_roads, family = 'NBP', method = 'NM')
                    
                    
comptable <- 
 regCompTable(list("NB-1"=nb.1, "NB-2"=nb.2, "NB-P"=nb.p), tableType="latex")
print(comptable)