R语言分析实战——考研人数的影响因素

原始数据

代码

data = read.csv('原始数据.csv',header = T)

# 画图
attach(data)
par(mfrow=c(2,2))
plot(年份, 报名人数,  pch = 15, lty = 1, col = "red", xlab = "年份", ylab = "报名人数")
abline(lsfit(年份,报名人数))
plot(年份, 录取人数, type = "b", pch = 16, lty = 2, col = "blue", xlab = "年份", ylab = "录取人数")
plot(年份, 毕业生人数, type = "b", pch = 17, lty = 3, col = "orange", xlab = "年份", ylab = "毕业生人数")
plot(年份, 就业率, type = "b", pch = 15, lty = 1, col = "black", xlab = "年份", ylab = "就业率")
par(mfrow=c(1,1))
detach(data)

# 相关性分析
#library(psych)
#corr.test(data[,c(2,3,4,5)])  # 上面是相关系数，下面是检测值
cov(data[,c(2,3,4,5)])  # 查看相关系数
# 报名人数与录取人数的相关性
cor.test(data[,2],data[,3])
# 报名人数与毕业生人数的相关性
cor.test(data[,2],data[,4])
# 报名人数与就业率的相关性
cor.test(data[,2],data[,5])

# 多元线性回归
fit = lm(报名人数~录取人数+毕业生人数+就业率)
summary(fit)  # 查看拟合效果
options(digits = 4)  # 保留四位数
coef(fit)  # 查看系数
# 画回归后对照图形
plot(年份, 报名人数, type = "b", pch = 15, lty = 1, col = "black", xlab = "年份", ylab = "报名人数",ylim = c(140,300))
lines(年份,回归数据,type = "b",pch = 16,lty=2,col = "red")
legend("topleft",c('源数据','回归数据'), 
       lty = c(1,2), pch = c(15,16),col=c('black','red'))