setwd("D:/data")  # 设置工作目录，根据实际情况修改路径
d3.5<-read.csv("C:\\Users\\86167\\Desktop\\ex3.5.csv",header=T)  # 读取数据到数据框d3.5
glm.logit<-glm(admit~gre+gpa+rank,family=binomial(link =logit),data=d3.5)
summary(glm.logit)

## 
## Call:
## glm(formula = admit ~ gre + gpa + rank, family = binomial(link = logit), 
##     data = d3.5)
## 
## Coefficients:
##              Estimate Std. Error z value Pr(>|z|)    
## (Intercept) -3.449548   1.132846  -3.045  0.00233 ** 
## gre          0.002294   0.001092   2.101  0.03564 *  
## gpa          0.777014   0.327484   2.373  0.01766 *  
## rank        -0.560031   0.127137  -4.405 1.06e-05 ***
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 
## (Dispersion parameter for binomial family taken to be 1)
## 
##     Null deviance: 499.98  on 399  degrees of freedom
## Residual deviance: 459.44  on 396  degrees of freedom
## AIC: 467.44
## 
## Number of Fisher Scoring iterations: 4

##gre,gpa,rank对应的P值都比较小，比较显著。admit=-3.4495+0.0023gra+0.7770gpa-0.5600rank

yp<-predict(glm.logit,data.frame(gre=380,gpa=3.61,rank=3)) 
p.fit<-exp(yp)/(1+exp(yp));p.fit 

##         1 
## 0.1895527

##当gre=380，gpa=3.61,rank=3时，估计admit=1的概率约为0.1896，可能性约为18.96%.

yp<-predict(glm.logit,data.frame(gre=800,gpa=4,rank=1)) 
p.fit<-exp(yp)/(1+exp(yp));p.fit

##         1 
## 0.7178136

##当gre=800，gpa=4,rank=1时，估计admit=1的概率约为0.7178，可能性约为71.78%.