Computes criterion distribution under null hypothesis for all contingency tables possible for a feature and a target.
distr_crit(target, feature, criterion = "ig", iter_limit = 200)
target | {0,1}-valued target vector. See Details. |
---|---|
feature | {0,1}-valued feature vector. See Details. |
criterion | criterion used for calculations of distribution.
See |
iter_limit | limit the number of calculated contingence matrices. If
|
An object of class criterion_distribution
.
both target
and feature
vectors may contain only 0
and 1.
target_feature <- create_feature_target(10, 375, 15, 600) distr_crit(target = target_feature[,1], feature = target_feature[,2])#> criterion pdf cdf #> [1,] 1.212755e-05 1.612726e-01 0.0000000 #> [2,] 3.412667e-05 1.608437e-01 0.1612726 #> [3,] 1.613452e-04 1.372196e-01 0.3221164 #> [4,] 2.343364e-04 1.352954e-01 0.4593359 #> [5,] 4.774248e-04 9.945760e-02 0.5946314 #> [6,] 6.230561e-04 9.512837e-02 0.6940890 #> [7,] 9.582756e-04 6.152186e-02 0.7892174 #> [8,] 1.215027e-03 5.520642e-02 0.8507392 #> [9,] 1.603902e-03 3.247796e-02 0.9059456 #> [10,] 2.031418e-03 2.597607e-02 0.9384236 #> [11,] 2.416404e-03 1.460524e-02 0.9643997 #> [12,] 3.103497e-03 9.658645e-03 0.9790049 #> [13,] 3.400145e-03 5.573369e-03 0.9886636 #> [14,] 4.480193e-03 2.730717e-03 0.9942369 #> [15,] 4.562127e-03 1.793697e-03 0.9969676 #> [16,] 5.912660e-03 4.825149e-04 0.9987613 #> [17,] 6.246461e-03 5.514752e-04 0.9992439 #> [18,] 7.466493e-03 1.071282e-04 0.9997953 #> [19,] 8.583769e-03 7.084924e-05 0.9999025 #> [20,] 9.244804e-03 1.928308e-05 0.9999733 #> [21,] 1.127887e-02 2.742703e-06 0.9999926 #> [22,] 1.235335e-02 4.350327e-06 0.9999953 #> [23,] 1.361763e-02 2.965966e-07 0.9999997 #> [24,] 1.634604e-02 2.290898e-08 1.0000000 #> [25,] 1.964559e-02 1.125550e-09 1.0000000 #> [26,] 2.437751e-02 2.642754e-11 1.0000000 #> attr(,"plot_data") #> unsort_criterion unsort_prob #> 0 1.235335e-02 4.350327e-06 #> 1 8.583769e-03 7.084924e-05 #> 2 6.246461e-03 5.514752e-04 #> 3 4.480193e-03 2.730717e-03 #> 4 3.103497e-03 9.658645e-03 #> 5 2.031418e-03 2.597607e-02 #> 6 1.215027e-03 5.520642e-02 #> 7 6.230561e-04 9.512837e-02 #> 8 2.343364e-04 1.352954e-01 #> 9 3.412667e-05 1.608437e-01 #> 10 1.212755e-05 1.612726e-01 #> 11 1.613452e-04 1.372196e-01 #> 12 4.774248e-04 9.945760e-02 #> 13 9.582756e-04 6.152186e-02 #> 14 1.603902e-03 3.247796e-02 #> 15 2.416404e-03 1.460524e-02 #> 16 3.400145e-03 5.573369e-03 #> 17 4.562127e-03 1.793697e-03 #> 18 5.912660e-03 4.825149e-04 #> 19 7.466493e-03 1.071282e-04 #> 20 9.244804e-03 1.928308e-05 #> 21 1.127887e-02 2.742703e-06 #> 22 1.361763e-02 2.965966e-07 #> 23 1.634604e-02 2.290898e-08 #> 24 1.964559e-02 1.125550e-09 #> 25 2.437751e-02 2.642754e-11 #> attr(,"nice_name") #> [1] "Information Gain" #> attr(,"class") #> [1] "criterion_distribution"