Skip to contents

Compute Information Value for Predictive Variables

Source: R/create_IV.R
create_IV.Rd

This function calculates the Information Value (IV) for the selected numeric predictor variables in the dataset, given a specified outcome variable. The Information Value provides a measure of the predictive power of each variable in relation to the outcome variable, which can be useful in feature selection for predictive modeling.

Usage

create_IV(
  data,
  predictors = NULL,
  outcome,
  bins = 5,
  siglevel = 0.05,
  exc_sig = FALSE,
  return = "plot"
)

Arguments

data

A Person Query dataset in the form of a data frame.

predictors

A character vector specifying the columns to be used as predictors. Defaults to NULL, where all numeric vectors in the data will be used as predictors.

outcome

String specifying the column name for a binary variable, containing only the values 1 or 0.

bins

Number of bins to use, defaults to 5.

siglevel

Significance level to use in comparing populations for the outcomes, defaults to 0.05

exc_sig

Logical value determining whether to exclude values where the p-value lies below what is set at siglevel. Defaults to FALSE, where p-value calculation does not happen altogether.

return

String specifying what to return. This must be one of the following strings:

  • "plot"

  • "summary"

  • "list"

  • "plot-WOE"

  • "IV"

See Value for more information.

Value

A different output is returned depending on the value passed to the return

argument:

  • "plot": 'ggplot' object. A bar plot showing the IV value of the top (maximum 12) variables.

  • "summary": data frame. A summary table for the metric.

  • "list": list. A list of outputs for all the input variables.

  • "plot-WOE": A list of 'ggplot' objects that show the WOE for each predictor used in the model.

  • "IV" returns a list object which mirrors the return in Information::create_infotables().

Details

This is a wrapper around wpa::create_IV().

See also

Other Variable Association: IV_report()

Other Information Value: IV_report()

Examples

# Return a summary table of IV
pq_data %>%
  dplyr::mutate(X = ifelse(Internal_network_size > 40, 1, 0)) %>%
  create_IV(outcome = "X",
            predictors = c("Email_hours",
                           "Meeting_hours",
                           "Chat_hours"),
            return = "plot")



# Return summary
pq_data %>%
  dplyr::mutate(X = ifelse(Internal_network_size > 40, 1, 0)) %>%
  create_IV(outcome = "X",
            predictors = c("Email_hours", "Meeting_hours"),
            return = "summary")
#>        Variable        IV
#> 1   Email_hours 1.7810969
#> 2 Meeting_hours 0.9445303