- 博客(11)
- 收藏
- 关注
RSS订阅原创 自助法和蒙特卡罗
##蒙特卡罗仿真：用发射随机数的方法模拟数据，根据模拟1000次之后得到的结果预测，由于随机数是正态分布的，要用mvrnorm方法模拟随机数。##勾MASS packagealpha=c()for(i in 1:1000){mu1=c(0,0)sigma1=matrix(c(1,0.5,0.5,1.25),nrow=2) ###σ²x=1, σ²y=1.25, and σxy=0...
2019-05-31 01:08:32
1027
原创 错误率曲线
lg.fit = glm(BAD~.,data = x,family = binomial) #family默认为gaussianerror1=c(rep(0,100)) #总错误的概率error2=c(rep(0,100)) #违约者被错误分类error3=c(rep(0,100)) #没有违约者被错误分类xx = seq(0.005,0.5,0.005)j=1for (i...
2019-05-31 00:47:30
677
原创 蒙特卡洛
library(MASS)n=100aplha = c()for (i in 1:10000){mu1=c(0,0) #均值是0sigma1=matrix(c(1,0.5,0.5,1.25),nrow =2)rand1=mvrnorm(n=100,mu =mu1,Sigma = sigma1)x=rand1[,1]y= rand1[,2]alpha[i]=(var(...
2019-05-31 00:44:39
1191
原创 ROC
d_train = rbind(d0_train,d1_train,d2_train,d3_train)d_test = rbind(d0_test,d1_test,d2_test,d3_test)d_trainy[dtrainy[d_trainy[dtrainy>=1]=1d_testy[dtesty[d_testy[dtesty>=1]=1d_trainy=as.fa...
2019-05-31 00:39:30
252
原创 LOOCV交叉验证题
#p是多少个变量x，p=1；n是100,记录了n行数据set.seed(1)y=rnorm(100)x=rnorm(100)y=x-2*x^2+rnorm(100)plot(x,y)##############
2019-05-31 00:25:33
1440
1
原创 PCA－主成分分析
PCA #主成分分析d=read.csv(“train.csv”,header=TRUE)View(d)dc=d[,2:9] sdc=scale(dc) #标准化mean(sdc[,1]) #检测有没有标准化成功cov_sdc=cov(sdc) #协方差eigen(cov_sdc) #princomp(sdc) &...
2019-05-30 23:46:00
509
原创 5.4bootstrap作业-qyc
We will now derive the probability that a given observation is part of a bootstrap sample. Suppose that we obtain a bootstrap sample from a set of n observations.(a) What is the probability that the...
2019-05-30 01:48:29
885
原创 project-随机森林-qyc
library(randomForest)d_train $ RainTomorrow= as.factor(d_train$RainTomorrow)err = matrix(rep(0,512),nrow = 5,byrow=TRUE)for (i in 1:5){for (j in 1:12){model = randomForest(d_train $ RainTomorrow~...
2019-05-30 00:43:49
203
原创 随机森林-ROC
install.packages(“randomForest”)library(randomForest)d0_train=d0[label0<=5,]d0_test=d0[label0>5,]d1_train=d1[label1<=5,]d1_test=d1[label1>5,]d2_train=d2[label2<=5,]d2_test=d2[lab...
2019-05-30 00:41:17
7323
原创 决策树-分数据集-建模预测选择决策树模型-剪枝-qyc
d = read.csv(“train.csv”,header = TRUE)dc= d[complete.cases(d),]#分类数据Y0~3 保持原来Y的分布，进行平均抽d0=d[dy==0,]d1=d[dy==0,]d1=d[dy==0,]d1=d[dy1,]d2=d[dy==2,]d3=d[dy==2,]d3=d[dy==2,]d3=d[dy3,]#自己分训练集和测试集l...
2019-05-29 21:55:49
509
原创 线性判别-qyc
setwd(“C:/Users/y/Desktop/r”) #路径 MOREd = read.csv(“hmeq.csv”,na.strings="")dc = d[complete.cases(d),] #只选取完整的行 选TRUE的行（因为，前）mdist = function(x){ #自定义函数 功能如下t = as.matrix(x) #变成矩阵p = dim...
2019-05-29 15:58:12
301
空空如也
空空如也
TA创建的收藏夹 TA关注的收藏夹
TA关注的人