README.md
In GabyP/preparation4modeling: What the Package Does (One Line, Title Case)
preparation4modelling
gabriela plantie
2019-01
Overview
This is a package that groups variables (categorical and numerical) according to their likelihood of producing the event analyzed. It can be used as a tool to explore information for building a model to predict a binary target.
Install
library(devtools)
#devtools::install_github("gabriela-plantie/preparation4modeling", force=T, dependencies = F)
library(preparation4modeling)
Usage
Example table
set.seed(1)
x1 = rnorm(1000)
x2 = rnorm(1000)
x4='A'
x4=ifelse(x1>0.1,'B', x4)
x4=ifelse(x1>0.4,'C', x4 )
x4=ifelse(x1>0.6,'D', x4 )
x4=ifelse(x1>0.8,'E', x4 )
z = 1 + 3*x1
pr = 1/(1+exp(-z))
y = rbinom(1000,1,pr)
tbla = data.frame(y=y,x1=x1,x2=x2, x4=x4)
q_nas=100
x1[1:q_nas] = NA
x4[1:q_nas]=NA
Analyzing variables against target
numerical variable with tree
agrupa_ctree (tbla, target_name='y', variable_name='x1',flag_numerica=1, max_q_groups=10, algoritmo='chaid' )
#> variable_name nodo_pred rangos_pred cant_nodo pos_nodo rt_nodo
#> 1 x1 1 (-Inf,-1.025] 164 2 0.012
#> 2 x1 2 (-1.016,-0.779] 59 9 0.153
#> 3 x1 2 (-0.772,-0.404] 119 50 0.420
#> 4 x1 3 (-0.4,0.333] 282 200 0.709
#> 5 x1 4 (0.341,0.616] 102 88 0.863
#> 6 x1 5 (0.62,Inf] 274 271 0.989
#> participacion log_odds corte_inf corte_sup
#> 1 0.164 -4.411 -3.008 -1.025
#> 2 0.059 -1.711 -1.016 -0.779
#> 3 0.119 -0.323 -0.772 -0.404
#> 4 0.282 0.891 -0.400 0.333
#> 5 0.102 1.840 0.341 0.616
#> 6 0.274 4.499 0.620 3.810
categorical variable with tree
agrupa_ctree (tbla, target_name='y', variable_name='x4',flag_numerica=0, algoritmo='chaid' )
#> variable_name nodo_pred variable_valor cant_nodo pos_nodo rt_nodo
#> 1 x4 1 E 214 212 0.991
#> 2 x4 2 B 104 86 0.827
#> 3 x4 3 A 548 201 0.367
#> 5 x4 pocos_casos D 134 121 0.903
#> 4 x4 pocos_casos C 134 121 0.903
#> participacion log_odds cant_var pos_var rt_var
#> 1 0.214 4.701 214 212 0.991
#> 2 0.104 1.565 104 86 0.827
#> 3 0.548 -0.545 548 201 0.367
#> 5 0.134 2.231 64 61 0.953
#> 4 0.134 2.231 70 60 0.857
categorical variable with hipergeometric test
agrupa_nominal_filtra_small(tbla, target_name='y', variable_name='x4',limite=0.05, symbol_to_split='%#%', limite_grupo=100)
#> variable_name nodo_pred variable_valor cant_nodo pos_nodo rt_nodo
#> 3 x4 1 E 214 212 0.9910000
#> 2 x4 2 B 104 86 0.8270000
#> 1 x4 3 A 548 201 0.3670000
#> 11 x4 pocos_casos C 134 121 0.9029851
#> 21 x4 pocos_casos D 134 121 0.9029851
#> participacion log_odds cant_var pos_var rt_var
#> 3 0.247 4.701000 214 212 0.9906542
#> 2 0.120 1.565000 104 86 0.8269231
#> 1 0.633 -0.545000 548 201 0.3667883
#> 11 Inf 2.230841 70 60 0.8571429
#> 21 Inf 2.230841 64 61 0.9531250
example table 2
x1 = rnorm(1000)
x2 = rnorm(1000)
x3= ifelse(as.factor(x2>0.5)==T, 'A', 'B')
x4= ifelse(as.factor(x2>0.7)==T, 'C', 'D')
z = 1 + 2 * x1 + 3 * x2
pr = 1/(1+exp(-z))
y = rbinom(1000,1,pr)
tbla = data.frame(y=y,x1=x1,x2=x2, x3=x3, x4=x4)
define level according to bad rate
tbla<-redefine_level_0( df_agrupada_y=tbla ,variables=c('x3', 'x4') ,nombre_target='y')
#> [1] "x3"
#> [1] "x4"
generating model and scorecard table
filtros_train= (tbla$random=runif(nrow(tbla)))<0.5
f=formula(y~x3+x4)
lr <- glm(f, tbla[ filtros_train, ], family = 'binomial')
tabla_estimadores(lr)
#> $scorecard
#> variable num_variable variable_nivel nivel Estimate P_value
#> 1 Intercepto 0 <NA> 3.277144 1.217737e-10
#> 2 x3 1 x3A A 0.000000 NA
#> 3 x3 1 x3B B -1.808817 7.673659e-05
#> 4 x4 2 x4C C 0.000000 NA
#> 5 x4 2 x4D D -1.572396 1.990559e-02
#> signif max_estim importancia ranking puntos
#> 1 *** NA NA 1 0
#> 2 1.808817 0.535 2 0
#> 3 *** 1.808817 0.535 1 535
#> 4 1.572396 0.465 2 0
#> 5 *** 1.572396 0.465 1 465
#>
#> $multiplier
#> [1] -295.7518
#>
#> $theorical_sum
#> [1] 1000
#>
#> $real_sum
#> [1] 1000
example table 3
x1 = rnorm(1000)
x2 = rnorm(1000)
z = 1 + 2 * x1 + 3 * x2
pr = 1/(1+exp(-z))
y = rbinom(1000,1,pr)
y1 = rbinom(1000,1,abs(pr-0.05))
tbla = data.frame(y=y,x1=x1,x2=x2, y1=y1)
f=formula(y~x1+x2)
lr <- glm(f, tbla, family = 'binomial')
tbla$prob<-predict(lr, tbla, type='response')
generating performance table
ventiles(tbla, targets=c('y', 'y1'), score_name = 'prob')
#> grupos tot br_y br_y1 ks_y ks_y1 min_prob max_prob
#> 20 [0,0.007] 50 0.000 0.060 0.12 0.10 NA 0.0070
#> 1 (0.007,0.0319] 50 0.020 0.000 0.25 0.21 0.0070 0.0319
#> 2 (0.0319,0.0792] 50 0.020 0.020 0.37 0.32 0.0319 0.0792
#> 3 (0.0792,0.145] 50 0.160 0.060 0.45 0.42 0.0792 0.1450
#> 4 (0.145,0.218] 50 0.140 0.120 0.55 0.51 0.1450 0.2180
#> 5 (0.218,0.304] 50 0.280 0.100 0.62 0.61 0.2180 0.3040
#> 6 (0.304,0.415] 50 0.500 0.280 0.64 0.66 0.3040 0.4150
#> 7 (0.415,0.533] 50 0.320 0.400 0.70 0.69 0.4150 0.5330
#> 8 (0.533,0.636] 50 0.620 0.540 0.70 0.70 0.5330 0.6360
#> 9 (0.636,0.728] 50 0.620 0.600 0.69 0.68 0.6360 0.7280
#> 10 (0.728,0.793] 50 0.800 0.680 0.65 0.66 0.7280 0.7930
#> 11 (0.793,0.852] 50 0.820 0.800 0.60 0.61 0.7930 0.8520
#> 12 (0.852,0.9] 50 0.900 0.840 0.54 0.56 0.8520 0.9000
#> 13 (0.9,0.932] 50 0.900 0.920 0.48 0.48 0.9000 0.9320
#> 14 (0.932,0.957] 50 0.960 0.920 0.40 0.41 0.9320 0.9570
#> 15 (0.957,0.975] 50 0.960 0.960 0.33 0.32 0.9570 0.9750
#> 16 (0.975,0.984] 50 1.000 0.940 0.25 0.25 0.9750 0.9840
#> 17 (0.984,0.993] 50 0.960 0.960 0.17 0.16 0.9840 0.9930
#> 18 (0.993,0.997] 50 1.000 0.940 0.08 0.09 0.9930 0.9970
#> 19 (0.997,1.1] 50 1.000 0.980 0.00 0.00 0.9970 1.1000
#> 21 todos 1000 0.599 0.556 0.70 0.70 NA NA
GabyP/preparation4modeling documentation built on Sept. 24, 2020, 11:57 p.m.
R Package Documentation
Browse R Packages
We want your feedback!
Note that we can't provide technical support on individual packages. You should contact the package authors for that.
preparation4modelling
gabriela plantie 2019-01
Overview
This is a package that groups variables (categorical and numerical) according to their likelihood of producing the event analyzed. It can be used as a tool to explore information for building a model to predict a binary target.
Install
library(devtools)
#devtools::install_github("gabriela-plantie/preparation4modeling", force=T, dependencies = F)
library(preparation4modeling)
Usage
Example table
set.seed(1)
x1 = rnorm(1000)
x2 = rnorm(1000)
x4='A'
x4=ifelse(x1>0.1,'B', x4)
x4=ifelse(x1>0.4,'C', x4 )
x4=ifelse(x1>0.6,'D', x4 )
x4=ifelse(x1>0.8,'E', x4 )
z = 1 + 3*x1
pr = 1/(1+exp(-z))
y = rbinom(1000,1,pr)
tbla = data.frame(y=y,x1=x1,x2=x2, x4=x4)
q_nas=100
x1[1:q_nas] = NA
x4[1:q_nas]=NA
Analyzing variables against target
numerical variable with tree
agrupa_ctree (tbla, target_name='y', variable_name='x1',flag_numerica=1, max_q_groups=10, algoritmo='chaid' )
#> variable_name nodo_pred rangos_pred cant_nodo pos_nodo rt_nodo
#> 1 x1 1 (-Inf,-1.025] 164 2 0.012
#> 2 x1 2 (-1.016,-0.779] 59 9 0.153
#> 3 x1 2 (-0.772,-0.404] 119 50 0.420
#> 4 x1 3 (-0.4,0.333] 282 200 0.709
#> 5 x1 4 (0.341,0.616] 102 88 0.863
#> 6 x1 5 (0.62,Inf] 274 271 0.989
#> participacion log_odds corte_inf corte_sup
#> 1 0.164 -4.411 -3.008 -1.025
#> 2 0.059 -1.711 -1.016 -0.779
#> 3 0.119 -0.323 -0.772 -0.404
#> 4 0.282 0.891 -0.400 0.333
#> 5 0.102 1.840 0.341 0.616
#> 6 0.274 4.499 0.620 3.810
categorical variable with tree
agrupa_ctree (tbla, target_name='y', variable_name='x4',flag_numerica=0, algoritmo='chaid' )
#> variable_name nodo_pred variable_valor cant_nodo pos_nodo rt_nodo
#> 1 x4 1 E 214 212 0.991
#> 2 x4 2 B 104 86 0.827
#> 3 x4 3 A 548 201 0.367
#> 5 x4 pocos_casos D 134 121 0.903
#> 4 x4 pocos_casos C 134 121 0.903
#> participacion log_odds cant_var pos_var rt_var
#> 1 0.214 4.701 214 212 0.991
#> 2 0.104 1.565 104 86 0.827
#> 3 0.548 -0.545 548 201 0.367
#> 5 0.134 2.231 64 61 0.953
#> 4 0.134 2.231 70 60 0.857
categorical variable with hipergeometric test
agrupa_nominal_filtra_small(tbla, target_name='y', variable_name='x4',limite=0.05, symbol_to_split='%#%', limite_grupo=100)
#> variable_name nodo_pred variable_valor cant_nodo pos_nodo rt_nodo
#> 3 x4 1 E 214 212 0.9910000
#> 2 x4 2 B 104 86 0.8270000
#> 1 x4 3 A 548 201 0.3670000
#> 11 x4 pocos_casos C 134 121 0.9029851
#> 21 x4 pocos_casos D 134 121 0.9029851
#> participacion log_odds cant_var pos_var rt_var
#> 3 0.247 4.701000 214 212 0.9906542
#> 2 0.120 1.565000 104 86 0.8269231
#> 1 0.633 -0.545000 548 201 0.3667883
#> 11 Inf 2.230841 70 60 0.8571429
#> 21 Inf 2.230841 64 61 0.9531250
example table 2
x1 = rnorm(1000)
x2 = rnorm(1000)
x3= ifelse(as.factor(x2>0.5)==T, 'A', 'B')
x4= ifelse(as.factor(x2>0.7)==T, 'C', 'D')
z = 1 + 2 * x1 + 3 * x2
pr = 1/(1+exp(-z))
y = rbinom(1000,1,pr)
tbla = data.frame(y=y,x1=x1,x2=x2, x3=x3, x4=x4)
define level according to bad rate
tbla<-redefine_level_0( df_agrupada_y=tbla ,variables=c('x3', 'x4') ,nombre_target='y')
#> [1] "x3"
#> [1] "x4"
generating model and scorecard table
filtros_train= (tbla$random=runif(nrow(tbla)))<0.5
f=formula(y~x3+x4)
lr <- glm(f, tbla[ filtros_train, ], family = 'binomial')
tabla_estimadores(lr)
#> $scorecard
#> variable num_variable variable_nivel nivel Estimate P_value
#> 1 Intercepto 0 <NA> 3.277144 1.217737e-10
#> 2 x3 1 x3A A 0.000000 NA
#> 3 x3 1 x3B B -1.808817 7.673659e-05
#> 4 x4 2 x4C C 0.000000 NA
#> 5 x4 2 x4D D -1.572396 1.990559e-02
#> signif max_estim importancia ranking puntos
#> 1 *** NA NA 1 0
#> 2 1.808817 0.535 2 0
#> 3 *** 1.808817 0.535 1 535
#> 4 1.572396 0.465 2 0
#> 5 *** 1.572396 0.465 1 465
#>
#> $multiplier
#> [1] -295.7518
#>
#> $theorical_sum
#> [1] 1000
#>
#> $real_sum
#> [1] 1000
example table 3
x1 = rnorm(1000)
x2 = rnorm(1000)
z = 1 + 2 * x1 + 3 * x2
pr = 1/(1+exp(-z))
y = rbinom(1000,1,pr)
y1 = rbinom(1000,1,abs(pr-0.05))
tbla = data.frame(y=y,x1=x1,x2=x2, y1=y1)
f=formula(y~x1+x2)
lr <- glm(f, tbla, family = 'binomial')
tbla$prob<-predict(lr, tbla, type='response')
generating performance table
ventiles(tbla, targets=c('y', 'y1'), score_name = 'prob')
#> grupos tot br_y br_y1 ks_y ks_y1 min_prob max_prob
#> 20 [0,0.007] 50 0.000 0.060 0.12 0.10 NA 0.0070
#> 1 (0.007,0.0319] 50 0.020 0.000 0.25 0.21 0.0070 0.0319
#> 2 (0.0319,0.0792] 50 0.020 0.020 0.37 0.32 0.0319 0.0792
#> 3 (0.0792,0.145] 50 0.160 0.060 0.45 0.42 0.0792 0.1450
#> 4 (0.145,0.218] 50 0.140 0.120 0.55 0.51 0.1450 0.2180
#> 5 (0.218,0.304] 50 0.280 0.100 0.62 0.61 0.2180 0.3040
#> 6 (0.304,0.415] 50 0.500 0.280 0.64 0.66 0.3040 0.4150
#> 7 (0.415,0.533] 50 0.320 0.400 0.70 0.69 0.4150 0.5330
#> 8 (0.533,0.636] 50 0.620 0.540 0.70 0.70 0.5330 0.6360
#> 9 (0.636,0.728] 50 0.620 0.600 0.69 0.68 0.6360 0.7280
#> 10 (0.728,0.793] 50 0.800 0.680 0.65 0.66 0.7280 0.7930
#> 11 (0.793,0.852] 50 0.820 0.800 0.60 0.61 0.7930 0.8520
#> 12 (0.852,0.9] 50 0.900 0.840 0.54 0.56 0.8520 0.9000
#> 13 (0.9,0.932] 50 0.900 0.920 0.48 0.48 0.9000 0.9320
#> 14 (0.932,0.957] 50 0.960 0.920 0.40 0.41 0.9320 0.9570
#> 15 (0.957,0.975] 50 0.960 0.960 0.33 0.32 0.9570 0.9750
#> 16 (0.975,0.984] 50 1.000 0.940 0.25 0.25 0.9750 0.9840
#> 17 (0.984,0.993] 50 0.960 0.960 0.17 0.16 0.9840 0.9930
#> 18 (0.993,0.997] 50 1.000 0.940 0.08 0.09 0.9930 0.9970
#> 19 (0.997,1.1] 50 1.000 0.980 0.00 0.00 0.9970 1.1000
#> 21 todos 1000 0.599 0.556 0.70 0.70 NA NA
R Package Documentation
Browse R Packages
We want your feedback!
Note that we can't provide technical support on individual packages. You should contact the package authors for that.
Embedding an R snippet on your website
Add the following code to your website.
For more information on customizing the embed code, read Embedding Snippets.