【R语言 评分模型】R语言建立信用评分模型

最新推荐文章于 2024-04-19 09:16:51 发布
置顶 最新推荐文章于 2024-04-19 09:16:51 发布
阅读量6.5k 收藏 63
点赞数 1
分类专栏： 数据科学--机器学习 文章标签： r语言
版权声明：本文为博主原创文章，遵循 CC 4.0 BY-SA 版权协议，转载请附上原文出处链接和本声明。
本文链接：https://blog.csdn.net/u013421629/article/details/78416067
版权

1、数据源：
我们将会使用在信用评级建模中非常常用的德国信贷数据（German credit dataset）作为建模的数据集。德国信贷数据共有1000条数据，每条数据20个特征。

2、数据源下载：
https://github.com/frankhlchi/R-scorecard

3、建模过程

这里写图片描述

4、完整版（源代码）：

rm(list=ls())
gc()
library(caret)
library(smbinning)
library(ggplot2)
library(woe)

#load the data
german_credit <- read.csv("C:/pic/credit/german_credit.csv",sep = ",")
train <-createDataPartition(y=german_credit$Creditability,p=0.75,list=FALSE)
train2 <- german_credit[train, ]
test2 <- german_credit[-train, ]

#Explore data distribution 
ggplot(german_credit, aes(x = Duration,y = ..count..,)) + geom_histogram(fill = "blue", colour = "grey60", size = 0.2, alpha = 0.2,binwidth = 5)
ggplot(german_credit, aes(x = CreditAmount,y = ..count..,)) + geom_histogram(fill = "blue", colour = "grey60", size = 0.2, alpha = 0.2,binwidth = 1000)
ggplot(german_credit, aes(x = Age,y = ..count..,)) + geom_histogram(fill = "blue", colour = "grey60", size = 0.2, alpha = 0.2,binwidth = 5)
ggplot(german_credit, aes(x =Creditability,y = ..count..,)) + geom_histogram(fill = "blue", colour = "grey60" , alpha = 0.2,binwidth = 0.5)

#Optimal Binning
Durationresult=smbinning(df=train2,y="Creditability",x="Duration",p=0.05)
CreditAmountresult=smbinning(df=train2,y="Creditability",x="CreditAmount",p=0.05) 
Ageresult=smbinning(df=train2,y="Creditability",x="Age",p=0.05) 
smbinning.plot(CreditAmountresult,option="WoE",sub="CreditAmount") 
smbinning.plot(Durationresult,option="WoE",sub="Duration")
smbinning.plot(Ageresult,option="WoE",sub="Age")



#Univariate Analysis
AccountBalancewoe=woe(train2, "AccountBalance",Continuous = F, "Creditability",C_Bin = 4,Good = "1",Bad = "0")
ggplot(AccountBalancewoe, aes(x = BIN, y = -WOE)) + geom_bar(stat = "identity",fill = "blue", colour = "grey60",size = 0.2, alpha = 0.2)+labs(title = "AccountBalance") 
ValueSavingswoe=woe(train2, "ValueSavings",Continuous = F, "Creditability",C_Bin = 5,Good = "1",Bad = "0")
ggplot(ValueSavingswoe, aes(x = BIN, y = -WOE)) + geom_bar(stat = "identity",fill = "blue", colour = "grey60",size = 0.2, alpha = 0.2)+labs(title = "ValueSavings") 
Lengthofcurrentemploymentwoe=woe(train2, "Lengthofcurrentemployment",Continuous = F, "Creditability",C_Bin = 4,Good = "1",Bad = "0")
ggplot(Lengthofcurrentemploymentwoe, aes(x = BIN, y = -WOE)) + geom_bar(stat = "identity",fill = "blue", colour = "grey60",size = 0.2, alpha = 0.2)+labs(title = "Lengthofcurrentemployment") 

#combine some bins
for(i in 1:750){
  if(train2$ValueSavings[i]==1){train2$ValueSavings[i]=2}
}
for(i in 1:750){
  if(train2$Lengthofcurrentemployment[i]==5){train2$Lengthofcurrentemployment[i]=4}
}

library(corrplot)
#correlation analysis
cor1<-cor(train)
corrplot(cor1,tl.cex = 0.5)

#Infomation Value calculation
for(i in 1:1000){
  if(german_credit$Duration[i]<=11){german_credit$Duration[i]=1}
  else if(german_credit$Duration[i]<=30){german_credit$Duration[i]=2}
  else{german_credit$Duration[i]=3}
}

for(i in 1:1000){
  if(german_credit$Age[i]<=25){german_credit$Age[i]=1}
  else{german_credit$Age[i]=2}
}

for(i in 1:1000){
  if(german_credit$CreditAmount[i]<=6742){german_credit$CreditAmount[i]=1}
  else{german_credit$CreditAmount[i]=2}
}


for(i in 1:1000){
  if(german_credit$ValueSavings[i]==1){german_credit$ValueSavings[i]=2}
}

for(i in 1:1000){
  if(german_credit$Lengthofcurrentemployment[i]==5){german_credit$Lengthofcurrentemployment[i]=4}
}

AccountBalancewoe=woe(train2, "AccountBalance",Continuous = F, "Creditability",C_Bin = 4,Good = "1",Bad = "0")
Durationwoe=woe(train2, "Duration",Continuous = F, "Creditability",C_Bin = 2,Good = "1",Bad = "0")
PaymentStatusofPreviousCreditwoe=woe(train2, "PaymentStatusofPreviousCredit",Continuous = F, "Creditability",C_Bin = 4,Good = "1",Bad = "0")
Purposewoe = woe(train2, "Purpose",Continuous = F, "Creditability",C_Bin = 11,Good = "1",Bad = "0")
CreditAmountwoe= woe(train2, "CreditAmount",Continuous = F, "Creditability",C_Bin = 2,Good = "1",Bad = "0")
ValueSavingswoe =woe(train2, "ValueSavings",Continuous = F, "Creditability",C_Bin = 4,Good = "1",Bad = "0")
Lengthofcurrentemploymentwoe=woe(train2, "Lengthofcurrentemployment",Continuous = F, "Creditability",C_Bin = 4,Good = "1",Bad = "0")
Instalmentpercenwoet=woe(train2, "Instalmentpercent",Continuous = F, "Creditability",C_Bin = 4,Good = "1",Bad = "0")
Sex.Marital.Statuswoe=woe(train2, "Sex.Marital.Status",Continuous = F, "Creditability",C_Bin = 4,Good = "1",Bad = "0")
Guarantorswoe=woe(train2, "Guarantors",Continuous = F, "Creditability",C_Bin = 2,Good = "1",Bad = "0")
DurationinCurrentaddresswoe=woe(train2, "DurationinCurrentaddress",Continuous = F, "Creditability",C_Bin = 4,Good = "1",Bad = "0")
Mostvaluableavailableassetwoe=woe(train2, "Mostvaluableavailableasset",Continuous = F, "Creditability",C_Bin = 4,Good = "1",Bad = "0")
Agewoe=woe(train2, "Age",Continuous = F, "Creditability",C_Bin = 2,Good = "1",Bad = "0")
ConcurrentCreditswoe=woe(train2, "ConcurrentCredits",Continuous = F, "Creditability",C_Bin = 3,Good = "1",Bad = "0")
Typeofapartmentwoe=woe(train2, "Typeofapartment",Continuous = F, "Creditability",C_Bin = 3,Good = "1",Bad = "0")
NoofCreditatthisBankwoe=woe(train2, "NoofCreditatthisBank",Continuous = F, "Creditability",C_Bin = 3,Good = "1",Bad = "0")
Occupationwoe=woe(train2, "Occupation",Continuous = F, "Creditability",C_Bin = 4,Good = "1",Bad = "0")
Noofdependentswoe=woe(train2, "Noofdependents",Continuous = F, "Creditability",C_Bin = 2,Good = "1",Bad = "0")
Telephonewoe=woe(train2, "Telephone",Continuous = F, "Creditability",C_Bin = 2,Good = "1",Bad = "0")
ForeignWorkerwoe=woe(train2, "ForeignWorker",Continuous = F, "Creditability",C_Bin = 2,Good = "1",Bad = "0")

va = c("AccountBalance",  "Duration",   "PaymentStatusofPreviousCredit",    "Purpose",  "CreditAmount", "ValueSavings", "Lengthofcurrentemployment",    "Instalmentpercent","Sex.Marital.Status","Guarantors","DurationinCurrentaddress",   "Mostvaluableavailableasset",   "Age","ConcurrentCredits","Typeofapartment",    "NoofCreditatthisBank","Occupation","Noofdependents",   "Telephone" ,"ForeignWorker")
iv=c(sum(AccountBalancewoe$IV),sum(Durationwoe$IV),sum(PaymentStatusofPreviousCreditwoe$IV),sum(Purposewoe$IV),sum(CreditAmountwoe$IV),sum(ValueSavingswoe$IV),sum(Lengthofcurrentemploymentwoe$IV) ,sum(Instalmentpercenwoet$IV) ,sum(Sex.Marital.Statuswoe$IV) ,sum(Guarantorswoe$IV)  ,sum(DurationinCurrentaddresswoe$IV) ,sum(Mostvaluableavailableassetwoe$IV),sum(Agewoe$IV),sum(ConcurrentCreditswoe$IV),sum(Typeofapartmentwoe$IV),sum(NoofCreditatthisBankwoe$IV),sum(Occupationwoe$IV), sum(Noofdependentswoe$IV), sum(Telephonewoe$IV),sum(ForeignWorkerwoe$IV))
infovalue = data.frame(va,iv)
ggplot(infovalue, aes(x = va, y = iv)) + geom_bar(stat = "identity",fill = "blue", colour = "grey60",size = 0.2, alpha = 0.2)+labs(title = "Information value")+ theme(axis.text.x=element_text(angle=90,colour="black",size=10));

#WoE transformation
german_credit$DurationinCurrentaddress=NULL
german_credit$Guarantors=NULL
german_credit$Instalmentpercent=NULL
german_credit$NoofCreditatthisBank=NULL
german_credit$Occupation=NULL
german_credit$Noofdependents=NULL
german_credit$Telephone=NULL
AccountBalancewoe=woe(train2, "AccountBalance",Continuous = F, "Creditability",C_Bin = 4,Good = "1",Bad = "0")
Durationwoe=woe(train2, "Duration",Continuous = F, "Creditability",C_Bin = 2,Good = "1",Bad = "0")
PaymentStatusofPreviousCreditwoe=woe(train2, "PaymentStatusofPreviousCredit",Continuous = F, "Creditability",C_Bin = 4,Good = "1",Bad = "0")
Purposewoe = woe(train2, "Purpose",Continuous = F, "Creditability",C_Bin = 11,Good = "1",Bad = "0")
CreditAmountwoe= woe(train2, "CreditAmount",Continuous = F, "Creditability",C_Bin = 2,Good = "1",Bad = "0")
ValueSavingswoe =woe(train2, "ValueSavings",Continuous = F, "Creditability",C_Bin = 4,Good = "1",Bad = "0")
Lengthofcurrentemploymentwoe=woe(train2, "Lengthofcurrentemployment",Continuous = F, "Creditability",C_Bin = 4,Good = "1",Bad = "0")
Sex.Marital.Statuswoe=woe(train2, "Sex.Marital.Status",Continuous = F, "Creditability",C_Bin = 4,Good = "1",Bad = "0")
Mostvaluableavailableassetwoe=woe(train2, "Mostvaluableavailableasset",Continuous = F, "Creditability",C_Bin = 4,Good = "1",Bad = "0")
Agewoe=woe(train2, "Age",Continuous = F, "Creditability",C_Bin = 2,Good = "1",Bad = "0")
ConcurrentCreditswoe=woe(train2, "ConcurrentCredits",Continuous = F, "Creditability",C_Bin = 3,Good = "1",Bad = "0")
Typeofapartmentwoe=woe(train2, "Typeofapartment",Continuous = F, "Creditability",C_Bin = 3,Good = "1",Bad = "0")
ForeignWorkerwoe=woe(train2, "ForeignWorker",Continuous = F, "Creditability",C_Bin = 2,Good = "1",Bad = "0")

for(i in 1:1000){

  for(s in 1:4){
    if(german_credit$AccountBalance[i]==s){
      german_credit$AccountBalance[i]=-AccountBalancewoe$WOE[s]
    }
  }

  for(s in 1:3){
    if(german_credit$Duration[i]==s){
      german_credit$Duration[i]=-Durationwoe$WOE[s]
    }
  }

  for(s in 0:4){
    if(german_credit$PaymentStatusofPreviousCredit[i]==s){
      german_credit$PaymentStatusofPreviousCredit[i]=-PaymentStatusofPreviousCreditwoe$WOE[s+1]
    }
  }

  for(s in 0:10){
    if(s<=6){
      if(german_credit$Purpose[i]==s){
        german_credit$Purpose[i]=-Purposewoe$WOE[s+1]
      }
    }else{
      if(german_credit$Purpose[i]==s){
        german_credit$Purpose[i]=-Purposewoe$WOE[s]
      }
    }
  }

  for(s in 1:2){
    if(german_credit$CreditAmount[i]==s){
      german_credit$CreditAmount[i]=-CreditAmountwoe$WOE[s]
    }
  }

  for(s in 2:5){
    if(german_credit$ValueSavings[i]==s){
      german_credit$ValueSavings[i]=-ValueSavingswoe$WOE[s-1]
    }
  }

  for(s in 1:5){
    if(german_credit$Lengthofcurrentemployment[i]==s){
      german_credit$Lengthofcurrentemployment[i]=-Lengthofcurrentemploymentwoe$WOE[s]
    }
  }

  for(s in 1:5){
    if(german_credit$Sex.Marital.Status[i]==s){
      german_credit$Sex.Marital.Status[i]=-Sex.Marital.Statuswoe$WOE[s]
    }
  }

  for(s in 1:4){
    if(german_credit$Mostvaluableavailableasset[i]==s){
      german_credit$Mostvaluableavailableasset[i]=-Mostvaluableavailableassetwoe$WOE[s]
    }
  }

  for(s in 1:2){
    if(german_credit$Age[i]==s){
      german_credit$Age[i]=-Agewoe$WOE[s]
    }
  }

  for(s in 1:5){
    if(german_credit$ConcurrentCredits[i]==s){
      german_credit$ConcurrentCredits[i]=-ConcurrentCreditswoe$WOE[s]
    }
  }

  for(s in 1:5){
    if(german_credit$Typeofapartment[i]==s){
      german_credit$Typeofapartment[i]=-Typeofapartmentwoe$WOE[s]
    }
  }

  for(s in 1:2){
    if(german_credit$ForeignWorker[i]==s){
      german_credit$ForeignWorker[i]=-ForeignWorkerwoe$WOE[s]
    }
  }
}

#Stepwise regression & Logistic model buiding
fit<-glm(Creditability~ AccountBalance + Duration +PaymentStatusofPreviousCredit +Purpose + CreditAmount + ValueSavings + Lengthofcurrentemployment +Sex.Marital.Status+ Mostvaluableavailableasset + Age + ConcurrentCredits + Typeofapartment + ForeignWorker,train2,family = "binomial")
backwards = step(fit)
summary(backwards)
fit2<-glm(Creditability~ AccountBalance + Duration +PaymentStatusofPreviousCredit +Purpose + CreditAmount + ValueSavings + Lengthofcurrentemployment + Age + ConcurrentCredits  + ForeignWorker,train2,family = "binomial")
summary(fit2)
fit3<-glm(Creditability~ AccountBalance + Duration +PaymentStatusofPreviousCredit +Purpose + CreditAmount + ValueSavings + Lengthofcurrentemployment + Age  + ForeignWorker,train2,family = "binomial")
summary(fit3)

#VIF testing
library(car)
vif(fit3, digits =3)

#producing confusion matrix
prediction <- predict(fit3,newdata=test2)
for (i in 1:250) {
  if(prediction[i]>0.99){
    prediction[i]=1}
  else
  {prediction[i]=0}
}
confusionMatrix(prediction, test2$Creditability)

coe = (fit3$coefficients)
p <- 20/log(2)
q <- 600-20*log(2.5)/log(2)

base <- q + p*as.numeric(coe[1])

#score for AccountBalance
AccountBalanceSCORE = p*as.numeric(coe[2])*AccountBalancewoe$woe[1]*-1

p*as.numeric(coe[2])*AccountBalancewoe$WOE[1]*-1
p*as.numeric(coe[2])*AccountBalancewoe$WOE[2]*-1
p*as.numeric(coe[2])*AccountBalancewoe$WOE[3]*-1
p*as.numeric(coe[2])*AccountBalancewoe$WOE[4]*-1

#score for Duration
p*as.numeric(coe[3])*Durationwoe$WOE[1]*-1
p*as.numeric(coe[3])*Durationwoe$WOE[2]*-1
p*as.numeric(coe[3])*Durationwoe$WOE[3]*-1

#score for PaymentStatusofPreviousCredit
p*as.numeric(coe[4])*PaymentStatusofPreviousCreditwoe$WOE[1]*-1
p*as.numeric(coe[4])*PaymentStatusofPreviousCreditwoe$WOE[2]*-1
p*as.numeric(coe[4])*PaymentStatusofPreviousCreditwoe$WOE[3]*-1
p*as.numeric(coe[4])*PaymentStatusofPreviousCreditwoe$WOE[4]*-1
p*as.numeric(coe[4])*PaymentStatusofPreviousCreditwoe$WOE[5]*-1

#score for purpose 
for(i in 1:10){
  print(p*as.numeric(coe[5])*Purposewoe$WOE[i])*-1
}

p*as.numeric(coe[6])*CreditAmountwoe$WOE[1]*-1
p*as.numeric(coe[6])*CreditAmountwoe$WOE[2]*-1

p*as.numeric(coe[7])*ValueSavingswoe$WOE[1]*-1
p*as.numeric(coe[7])*ValueSavingswoe$WOE[2]*-1
p*as.numeric(coe[7])*ValueSavingswoe$WOE[3]*-1
p*as.numeric(coe[7])*ValueSavingswoe$WOE[4]*-1

p*as.numeric(coe[8])*Lengthofcurrentemploymentwoe$WOE[1]*-1
p*as.numeric(coe[8])*Lengthofcurrentemploymentwoe$WOE[2]*-1
p*as.numeric(coe[8])*Lengthofcurrentemploymentwoe$WOE[3]*-1
p*as.numeric(coe[8])*Lengthofcurrentemploymentwoe$WOE[4]*-1

p*as.numeric(coe[9])*Agewoe$WOE[1]*-1
p*as.numeric(coe[9])*Agewoe$WOE[2]*-1

p*as.numeric(coe[10])*ForeignWorkerwoe$WOE[1]*-1
p*as.numeric(coe[10])*ForeignWorkerwoe$WOE[2]*-1

建模方法总结篇：

1、变量分箱
在评分卡建模中，变量分箱（binning）是对连续变量离散化（discretization）的一种称呼。要将logistic模型转换为标准评分卡的形式，这一环节是必须完成的。信用评分卡开发中一般有常用的等距分段、等深分段、最优分段。

其中等距分段（Equval length intervals）是指分段的区间是一致的，比如年龄以十年作为一个分段；等深分段（Equal frequency intervals）是先确定分段数量，然后令每个分段中数据数量大致相等；最优分段（Optimal Binning）又叫监督离散化（supervised discretizaion），使用递归划分（Recursive Partitioning）将连续变量分为分段，背后是一种基于条件推断查找较佳分组的算法（Conditional Inference Tree）

2、单变量分析
在风险建模的过程中，变量选择可以具体细化为单变量变量筛选 (Univariate Variable Selection）和多变量变量筛选 (Multivariate Variable Selection)。多变量变量筛选一般会利用Stepwise算法在变量池中选取最优变量。 而单变量筛选，或者说单变量分析，是通过比较指标分箱和对应分箱的违约概率来确定指标是否符合经济意义。

3、相关性分析 & IV（信息值）筛选
我们会用经过清洗后的数据看一下变量间的相关性。注意，这里的相关性分析只是初步的检查，进一步检查模型的多重共线性还需要通过 VIF(variance inflation factor）也就是 方差膨胀因子进行检验。

4、StepWise多变量分析 & Logistic模型建立
在进行StepWise分析前，我们需要将筛选后的变量转换为WoE值并建立Logistic模型。
将经过WoE转换的数据放入Logistic模型中建模，并使用向后逐步回归方法
为防止多重共线性问题的出现，我们对模型进行VIF检验：

5、模型检验
到这里，我们的建模部分基本结束了。我们需要验证一下模型的预测能力如何。我们使用在建模开始阶段预留的250条数据进行检验:
模型的精度达到了0.72，模型表现一般。这同Logistic模型本身的局限性有关。传统的回归模型精度一般都会弱于决策树、SVM等机器挖掘算法。

6、打分卡转换
我们在上一部分，我们已经基本完成了建模相关的工作，并用混淆矩阵验证了模型的预测能力。接下来的步骤，就是将Logistic模型转换为标准打分卡的形式。
在建立标准评分卡之前，我们需要选取几个评分卡参数：基础分值、 PDO（比率翻倍的分值）和好坏比。 这里， 我们取600分为基础分值，PDO为20 （每高20分好坏比翻一倍），好坏比取2.5。；可得下式：

620 = q - p * log(2.5)
600 = q - p * log(2.5/2)
p = 20/log(2)
q =600-20*log(2.5)/log(2)
东华果汁哥
关注
  • 1
    点赞
  • 踩
  • 63
    收藏
    觉得还不错? 一键收藏
  • 打赏
    打赏
  • 2
    评论
专栏目录
基于python的评分卡模型代码实现
03-26
基于python的评分卡模型代码实现
应用预测建模第四章信用卡评分模型练习-R语言【不同的重抽样方法比较：重复K折交叉验证、K折交叉验证、留一交叉验证、重复训练/测试集划分 、Bootstrap；支持向量机与逻辑回归模型比较】
totobey的博客
11-24 2076
不同的重抽样方法比较：重复K折交叉验证、K交叉验证、LOOCV留一交叉验证、重复训练/测试集划分 （留多交叉验证、蒙特卡洛交叉验证）、Bootstrap自助法、632法（为消除估计偏差，Bootstrap自助法改进） 支持向量机与逻辑回归模型比较 应用预测建模 Applied Predictive Modeling (2013) by Max Kuhn and Kjell Johnson，林荟...
Python实验——用sklearn模块中的决策树方法对德国信贷数据进行分类。
weixin_45980783的博客
07-11 2026
一、简介 博主是一个普普通通的大学生，没有很厉害的技术，写的内容都是不太正经的偏小白简单的，写的也是学校教过的知识消化后自己的见解，不是很学术研究的博文。 本篇博文主要是关于Python扩展模块pandas中数据结构Series的创建及数据的访问，DataFrame的创建及数据的访问和关于Python的机器学习模块sklearn中决策树的构建。 配置：Window 7旗舰版+64位操作系统+Anaconda3 Spyder（64-bit） 这篇博文是博主的实验报告，之所以以图片形式上传就是避免抄袭。如果有
R语言逻辑回归(Logistic Regression)、回归决策树、随机森林信用卡违约分析信贷数据集...
拓端研究室TRL
08-11 86
原文链接：http://tecdat.cn/?p=23344在许多网站上都可以找到一个流行的德国信贷数据集_german_credit_，其中包含了银行贷款申请人的信息。该文件包含1000名申请人的20条信息。下面的代码可以用来确定申请人是否有信用，以及他（或她）是否对贷款人有良好的信用风险。有几种方法被应用到数据上，帮助做出这种判断。在这个案例中，我们将看一下这些方法。请注意，本例可能需要进行一...
德国信贷数据集GermanCredit
05-03
我们将会使用在信用评级建模中非常常用的德国信贷数据（German credit dataset）作为建模的数据集。德国信贷数据共有1000条数据，每条数据20个特征。这些特征包括AccountBalance（Checking账户余额）、Duration (Duration of Credit in month 借款期限）、Paymentstatus（还款记录）等。 其中比较难以理解的指标是Instalmentpercent，其代表着 Installment rate in percentage of disposable income （分期付款占可支配收入的百分比）。
Dataset之GermanCreditData：GermanCreditData数据集的简介、下载、使用方法之详细攻略
近期请国内外头部出版社可尽快私信博主！——心比天高，仗剑走天涯，保持热爱，奔赴向梦想！低调，谦虚，自律，反思，成长，还算是比较正能量的博主，公益免费传播……内心特别想在AI界做出一些可以推进历史进程影响力的东西(兴趣使然，有点小情怀，也有点使命感呀)…
06-23 4662
Dataset之GermanCreditData：GermanCreditData数据集的简介、下载、使用方法之详细攻略目录GermanCreditData数据集的简介1、数据集描述GermanCreditData数据集的下载GermanCreditData数据集的使用方法 德国信用数据，该数据集来自汉堡大学统计与计量研究所的教授，Hans Hofmann博士所收集制作。备注：DM指德国货币—马克，1DM≈11RMB字段英文中文描述7个数值型Duration in month月持续时间Cred
Logstic Regression模型对German Credit数据集进行分类
GODSuner的博客
03-20 2706
Case: German Credit 在这份作业中，我们使用了Logstic Regression模型对German Credit数据集进行了分类。并用混淆矩阵和ROC曲线对模型进行了评估。 若对本文存有疑问或获取数据代码，请直接私信博主或直接添加博主VX: 1178623893 The German Credit data set contains observations on 30 variables for 1000 past applicants for credit. Each app
利用R语言对贷款客户作风险评估
weixin_45670288的博客
12-30 3059
利用R语言对贷款客户作风险评估（上）——数据分析 前言 风险控制能力越来越成为互联网金融行业的隐形门槛，为风控人员提供显著地风险评估依据变得非常重要。本文以银行客户的信用卡信息为案例数据，对数据进行分析为信贷管理的风控人员提供风险评估依据。本文做以下几方面的工作： 处理原始数据，进行数据探索 好坏客户为因变量，处理不平衡数据的二分类问题 将逾期还款次数作为新的因变量做零膨胀回归问题，旨在为信贷管...
R语言使用GermanCredit数据集进行线性核函数SVM
Mrrunsen的博客
09-02 331
以德国信用数据为例。该数据几乎全是数字，并且大部分都是二进制变量(0/1)。正适合建立支持向量机, 或则随机森林。先从最简单的线性核函数开始, 需要注意的是，支持向量机建模前必须对数据归一化处理。正确率达到了77%， 并且在预测集上比训练集表现更好。这样正是SVM的优势。
R语言实现个人信用风险评估（数据科学导引）
m0_66291385的博客
02-24 464
在很多国家，政府机构会密切监控贷款业务银行需要明确解释申请者的贷款申请被拒绝或者批准的原因。这种可解释性对于贷款申请者也是很重要的，在贷款申请被银行拒绝时，申请者需要知道为什么自己的信用级别不符合银行的要求．通过构建自动化的信用评分模型，以在线方式进行即时的信贷审批能够为银行节约很多人工成本，数据集47 共有 1000个贷款样本，每个样本包含20 个相关特征和 1个表示贷款是否违约的特征“default” 。
R语言-基于广义线性模型的评分卡模型（数据集）
05-13
仅供学习使用
german_credit_data
12-27
德国信贷数据（German credit dataset）为信用评级建模中非常典型的数据。共有1000条数据，每条数据20个特征。本数据从github.com上下载的，原汁原味~
german credit data
11-29
互联网上经常被用来研究信用风险评级模型的加州大学机器学习数据库中的german credit data
GermanCredit:用于在 Azure ML 中探索和建模 UCI 德国信用数据集的 R 代码
05-31
使用 Azure 机器学习和 R 探索和分析 UCI 德国信用数据的 R 代码 此存储库包含R代码，用于使用和R探索和分析。
German Credit Data
10-13
German Credit Data 欺诈测试数据，从KAGGLE上下载的，用于反欺诈模型的训练数据
信用评分模型
04-28
信用评分模型 信用评分模型 信用评分模型 信用评分模型 信用评分模型 信用评分模型 信用评分模型 信用评分模型 信用评分模型 信用评分模型 信用评分模型 信用评分模型 信用评分模型 信用评分模型 信用评分模型 信用...
R语言 线性回归预测电影评分 报告
02-17
但是对于线性模型的R2非常的低，因此不能预测电影评分，需要再采集更多是数据进行预测电影评分，比如电影的剧情，可以做对应的编码处理。由于电影评分的影响因素有很多,因此再未来的研究中可以考虑在模型中加入更多...
信用卡评分模型源数据.rar
04-30
信用卡评分模型源数据
Relu激活函数
最新发布
一休哥的专栏
04-19 879
激活函数，relu的整体效果较好
R语言计算SVM模型的F1评分
06-09
计算SVM模型的F1评分可以使用R语言中的"caret"包。首先需要将数据集分为训练集和测试集，然后利用"train"函数训练SVM模型并预测测试集的结果。最后，使用"confusionMatrix"函数计算F1评分。 以下是示例代码： ```R # 导入所需包 library(caret) library(e1071) # 加载数据 data(iris) # 将数据集分为训练集和测试集 set.seed(123) trainIndex <- createDataPartition(iris$Species, p = .8, list = FALSE) training <- iris[trainIndex, ] testing <- iris[-trainIndex, ] # 训练SVM模型 model <- train(Species ~ ., data = training, method = "svmRadial") # 预测测试集的结果 predictions <- predict(model, newdata = testing) # 计算F1评分 cm <- confusionMatrix(predictions, testing$Species) f1_score <- cm$byClass["F1"] ``` 以上代码中，我们使用了"svmRadial"方法训练SVM模型。"confusionMatrix"函数返回的结果中，"byClass"列给出了每个类别的F1评分。

“相关推荐”对你有帮助么？

  • 非常没帮助
  • 没帮助
  • 一般
  • 有帮助
  • 非常有帮助
提交
评论 2
添加红包

请填写红包祝福语或标题

个

红包个数最小为10个

元

红包金额最低5元

当前余额3.43元 前往充值 >
需支付：10.00元
成就一亿技术人!
领取后你会自动成为博主和红包主的粉丝 规则
hope_wisdom
发出的红包

打赏作者

东华果汁哥

你的鼓励将是我创作的最大动力

¥1 ¥2 ¥4 ¥6 ¥10 ¥20
扫码支付：¥1
获取中
扫码支付

您的余额不足，请更换扫码支付或充值

打赏作者

实付元
使用余额支付
点击重新获取
扫码支付
钱包余额 0

抵扣说明：

1.余额是钱包充值的虚拟货币，按照1:1的比例进行支付金额的抵扣。
2.余额无法直接购买下载，可以购买VIP、付费专栏及课程。

余额充值