正文

R语言逻辑回归深入讲解

︶ㄣSDDLL学院 メ
SDDLL学院V管理员 / 阅读
0702
这篇文章主要介绍了R语言逻辑回归深入讲解，文中对于逻辑回归讲解的很细致，有感兴趣的同学可以研究下

这篇文章主要介绍了R语言逻辑回归深入讲解，文中对于逻辑回归讲解的很细致，有感兴趣的同学可以研究下

逻辑回归

?
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
> ###############逻辑回归
> setwd("/Users/yaozhilin/Downloads/R_edu/data")
> accepts<-read.csv("accepts.csv")
> names(accepts)
 [1] "application_id" "account_number" "bad_ind"    "vehicle_year"  "vehicle_make"
 [6] "bankruptcy_ind" "tot_derog"   "tot_tr"     "age_oldest_tr" "tot_open_tr" 
[11] "tot_rev_tr"   "tot_rev_debt"  "tot_rev_line"  "rev_util"    "fico_score" 
[16] "purch_price"  "msrp"      "down_pyt"    "loan_term"   "loan_amt"  
[21] "ltv"      "tot_income"   "veh_mileage"  "used_ind"  
> accepts<-accepts[complete.cases(accepts),]
> select<-sample(1:nrow(accepts),length(accepts$application_id)*0.7)
> train<-accepts[select,]###70%用于建模
> test<-accepts[-select,]###30%用于检测
> attach(train)
> ###用glm(y~x,family=binomial(link="logit"))
> gl<-glm(bad_ind~fico_score,family=binomial(link = "logit"))
> summary(gl)
 
Call:
glm(formula = bad_ind ~ fico_score, family = binomial(link = "logit"))
 
Deviance Residuals:
  Min    1Q  Median    3Q   Max
-2.0794 -0.6790 -0.4937 -0.3073  2.6028
 
Coefficients:
       Estimate Std. Error z value Pr(>|z|) 
(Intercept) 9.049667  0.629120  14.38  <2e-16 ***
fico_score -0.015407  0.000938 -16.43  <2e-16 ***
---
Signif. codes: 0 ‘***' 0.001 ‘**' 0.01 ‘*' 0.05 ‘.' 0.1 ‘ ' 1
 
(Dispersion parameter for binomial family taken to be 1)
 
  Null deviance: 2989.2 on 3046 degrees of freedom
Residual deviance: 2665.9 on 3045 degrees of freedom
AIC: 2669.9
 
Number of Fisher Scoring iterations: 5

多元逻辑回归

?
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
> ###多元逻辑回归
> gls<-glm(bad_ind~fico_score+bankruptcy_ind+age_oldest_tr+
+      tot_derog+rev_util+veh_mileage,family = binomial(link = "logit"))
> summary(gls)
 
Call:
glm(formula = bad_ind ~ fico_score + bankruptcy_ind + age_oldest_tr +
  tot_derog + rev_util + veh_mileage, family = binomial(link = "logit"))
 
Deviance Residuals:
  Min    1Q  Median    3Q   Max
-2.2646 -0.6743 -0.4647 -0.2630  2.8177
 
Coefficients:
         Estimate Std. Error z value Pr(>|z|) 
(Intercept)   8.205e+00 7.433e-01 11.039 < 2e-16 ***
fico_score   -1.338e-02 1.092e-03 -12.260 < 2e-16 ***
bankruptcy_indY -3.771e-01 1.855e-01 -2.033  0.0421 *
age_oldest_tr  -4.458e-03 6.375e-04 -6.994 2.68e-12 ***
tot_derog    3.012e-02 1.552e-02  1.941  0.0523 .
rev_util     3.763e-04 5.252e-04  0.717  0.4737 
veh_mileage   2.466e-06 1.381e-06  1.786  0.0741 .
---
Signif. codes: 0 ‘***' 0.001 ‘**' 0.01 ‘*' 0.05 ‘.' 0.1 ‘ ' 1
 
(Dispersion parameter for binomial family taken to be 1)
 
  Null deviance: 2989.2 on 3046 degrees of freedom
Residual deviance: 2601.4 on 3040 degrees of freedom
AIC: 2615.4
 
Number of Fisher Scoring iterations: 5
 
> glss<-step(gls,direction = "both")
Start: AIC=2615.35
bad_ind ~ fico_score + bankruptcy_ind + age_oldest_tr + tot_derog +
  rev_util + veh_mileage
 
         Df Deviance  AIC
- rev_util    1  2601.9 2613.9
<none>        2601.3 2615.3
- veh_mileage   1  2604.4 2616.4
- tot_derog    1  2605.1 2617.1
- bankruptcy_ind 1  2605.7 2617.7
- age_oldest_tr  1  2655.9 2667.9
- fico_score   1  2763.8 2775.8
 
Step: AIC=2613.88
bad_ind ~ fico_score + bankruptcy_ind + age_oldest_tr + tot_derog +
  veh_mileage
 
         Df Deviance  AIC
<none>        2601.9 2613.9
- veh_mileage   1  2604.9 2614.9
+ rev_util    1  2601.3 2615.3
- tot_derog    1  2605.7 2615.7
- bankruptcy_ind 1  2606.1 2616.1
- age_oldest_tr  1  2656.9 2666.9
- fico_score   1  2773.2 2783.2
?
1
2
3
4
5
6
7
8
9
10
11
> #出来的数据是logit，我们需要转换
> train$pre<-predict(glss,train)
> #出来的数据是logit，我们需要转换
> train$pre<-predict(glss,train)
> summary(train$pre)
  Min. 1st Qu. Median  Mean 3rd Qu.  Max.
 -4.868 -2.421 -1.671 -1.713 -1.011  2.497
> train$pre_p<-1/(1+exp(-1*train$pre))
> summary(train$pre_p)
  Min. 1st Qu. Median  Mean 3rd Qu.  Max.
0.00763 0.08157 0.15823 0.19298 0.26677 0.92395
?
1
2
3
4
5
#逻辑回归不需要检测扰动项，但需要检测共线性
> library(car)
> vif(glss)
> fico_score bankruptcy_ind age_oldest_tr   tot_derog  veh_mileage
>1.271283    1.144846    1.075603    1.423850    1.003616

到此这篇关于R语言逻辑回归深入讲解的文章就介绍到这了,更多相关R语言逻辑回归内容请搜索米米素材网以前的文章或继续浏览下面的相关文章希望大家以后多多支持米米素材网！

原文链接：https://www.cnblogs.com/ye20190812/p/13925635.html