ProcessRAI {rai}

Summarising RAI Output

Description

Processes the output from the rai function. Requires dplyr, tibble, and ggplot2 packages.

Usage

plot_ntest_rS(rawSum)

plot_ntest_wealth(rawSum)

## S3 method for class 'rai'
predict(object, newdata = NULL, alpha = NULL,
  omega = NULL, ...)

## S3 method for class 'rai'
summary(object, ...)

Arguments

rawSum

processed version of rai summary stored as a tibble with correct column parsing.

object

an object of class rai; expected to be the list output from the rai function.

newdata

an optional data frame in which to look for variables with which to predict. If omitted, the fitted values are used.

alpha

level of procedure.

omega

return from rejecting a test in Alpha-Investing (<= alpha).

...

additional arguments affecting the summary or predict methods.

Value

A list which includes the following components:

plot_rS

plot of the change in r.squared over time (number of tests conducted).

plot_wealth

plot of the change in r.squared over time (number of tests conducted).

experts

summary of expert performance: number of features, number of rejections, order in which they were added to the expert list.

tests

table of number of times features were tested: how many features tested k times; which expert(s) conducted tests.

epochs

in which epochs were tests rejected and the corresponding rejection thresholds.

stats

summary statistics: number of tests, number of epochs, bound on percentage reduction in ESS by adding a single feature, number of passes through to features, final r.squared, cost of raiPlus (0 for rai).

options

options given to RAI: algorithm, searchType, poly, startDegree, r.

Examples

  data("CO2")
  theResponse = CO2$uptake
  theData = CO2[ ,-5]
  rai_out = rai(theData, theResponse)
  summary(rai_out)  # summary information including graphs
  predict(rai_out)  # fitted values from selected model