boot.md
In fairadapt: Fair Data Adaptation with Quantile Preservation

fairadaptBoot

{
  "type": "double",
  "attributes": {},
  "value": [-0.0405844]
}

{
  "type": "double",
  "attributes": {},
  "value": [0.00487013]
}

Code
  print(ran)
Output

  Call:
  fairadaptBoot(formula = y ~ ., prot.attr = "a", adj.mat = adj.mat, 
      train.data = train, test.data = test, keep.object = TRUE, 
      n.boot = 3L, seed = 202)

  Bootstrap repetitions: 3

  Adapting variables:
    y, x

  Based on protected attribute a

    AND

  Based on causal graph:
    a y x
  a 0 0 1
  y 0 0 0
  x 0 1 0

Code
  summary(ran)
Output

  Call:
  fairadaptBoot(formula = y ~ ., prot.attr = "a", adj.mat = adj.mat, 
      train.data = train, test.data = test, keep.object = TRUE, 
      n.boot = 3L, seed = 202)

  Bootstrap repetitions:      3
  Protected attribute:        a
  Protected attribute levels: 0, 1
  Adapted variables:          y, x

  Number of training samples: 100
  Number of test samples:     0
  Quantile method:            quant.method

  Randomness considered:      finsamp
  fairadapt objects saved:    TRUE

Code
  print(rto)
Output

  Call:
  fairadaptBoot(formula = y ~ ., prot.attr = "a", train.data = train, 
      test.data = test, top.ord = c("a", "x", "y"), n.boot = 3L, 
      seed = 202)

  Bootstrap repetitions: 3

  Adapting variables:
    x, y

  Based on protected attribute a

    AND

  Based on topological order:
    axy

Code
  summary(rto)
Output

  Call:
  fairadaptBoot(formula = y ~ ., prot.attr = "a", train.data = train, 
      test.data = test, top.ord = c("a", "x", "y"), n.boot = 3L, 
      seed = 202)

  Bootstrap repetitions:      3
  Protected attribute:        a
  Protected attribute levels: 0, 1

  Number of training samples: 100
  Number of test samples:     0
  Quantile method:            quant.method

  Randomness considered:      finsamp
  fairadapt objects saved:    FALSE

Code
  print(charmod)
Output

  Call:
  fairadaptBoot(formula = score ~ ., prot.attr = "gender", adj.mat = adj.mat, 
      train.data = uni, keep.object = TRUE, n.boot = 3L, seed = 203)

  Bootstrap repetitions: 3

  Adapting variables:
    edu, test, score

  Based on protected attribute gender

    AND

  Based on causal graph:
         gender edu test score
  gender      0   1    1     1
  edu         0   0    0     1
  test        0   0    0     1
  score       0   0    0     0

Code
  summary(charmod)
Output

  Call:
  fairadaptBoot(formula = score ~ ., prot.attr = "gender", adj.mat = adj.mat, 
      train.data = uni, keep.object = TRUE, n.boot = 3L, seed = 203)

  Bootstrap repetitions:      3
  Protected attribute:        gender
  Protected attribute levels: 0, 1
  Adapted variables:          edu, test, score

  Number of training samples: 1000
  Number of test samples:     0
  Quantile method:            quant.method

  Randomness considered:      finsamp
  fairadapt objects saved:    TRUE

Code
  print(mod)
Output

  Call:
  fairadaptBoot(formula = two_year_recid ~ ., prot.attr = "race", 
      adj.mat = adj.mat, train.data = train, test.data = test, 
      n.boot = 3, seed = 203)

  Bootstrap repetitions: 3

  Adapting variables:
    juv_fel_count, juv_misd_count, juv_other_count, priors_count, c_charge_degree, two_year_recid

  Based on protected attribute race

    AND

  Based on causal graph:
                  age sex juv_fel_count juv_misd_count juv_other_count priors_count c_charge_degree race two_year_recid
  age               0   0             1              1               1            1               1    0              1
  sex               0   0             1              1               1            1               1    0              1
  juv_fel_count     0   0             0              0               0            1               1    0              1
  juv_misd_count    0   0             0              0               0            1               1    0              1
  juv_other_count   0   0             0              0               0            1               1    0              1
  priors_count      0   0             0              0               0            0               1    0              1
  c_charge_degree   0   0             0              0               0            0               0    0              1
  race              0   0             1              1               1            1               1    0              1
  two_year_recid    0   0             0              0               0            0               0    0              0

Code
  summary(mod)
Output

  Call:
  fairadaptBoot(formula = two_year_recid ~ ., prot.attr = "race", 
      adj.mat = adj.mat, train.data = train, test.data = test, 
      n.boot = 3, seed = 203)

  Bootstrap repetitions:      3
  Protected attribute:        race
  Protected attribute levels: Non-White, White
  Adapted variables:          juv_fel_count, juv_misd_count, juv_other_count, priors_count, c_charge_degree, two_year_recid

  Number of training samples: 1000
  Number of test samples:     0
  Quantile method:            quant.method

  Randomness considered:      finsamp
  fairadapt objects saved:    FALSE