比数比计算通用方法

Question

我有一个包含两个变量的数据框架，性别和城镇(Df1)。我想计算性别的优势比(用female=1)，我想每个城镇都这样做，这样我最终得到的是Df1的三个赔率比。

我的实际数据集包含了更多的城镇，所以我想知道是否有一种更普遍的方法来做到这一点，而不是手动将观察到的数量输入到墓志铭：：oddsratio()？

谢谢!

起点(df)：

Df1 <- data.frame(gender=c("m","m","m","f","f","f","m","m","m","f","m","f","m","f","f","f","f","f","f","f"), town=c("ny","la","ny","la","ny","la","ny","la","ny","la","ny","la","ny","la","ma","ma","ma","ma","ma","ma"))

目前为止的代码：

library(epitools)
Df2 <- matrix(c(12,20,8,20),byrow=TRUE,ncol=2)
dimnames(Df2) <- list(Group=c("females","males"),MI=c("subtotal","total"))
oddsratio(Df2)

注意事项：赔率比(字面意思是两个赔率之间的比率)

假设每10名男性中有7名入院: p=0.7，q=1-0.7=0.3

假设每10名女性中有3名被录取: p=0.3，q=1-0,3=0.7

男性入学几率: 0.7/0.3=2.333 (被录取/不被录取)

女性入学比率: 0.3/0.7=0.429

入学比数比: OR=2.333/0.429=5.44，

男性被接纳的几率是女性的5.44倍。

Answer 1

像这样吗？

library(tidyverse)
Df1 <- data.frame(gender=c("m","m","m","f","f","f","m","m","m","f","m","f","m","f","f","f","f","f","f","f"), town=c("ny","la","ny","la","ny","la","ny","la","ny","la","ny","la","ny","la","ma","ma","ma","ma","ma","ma"))

Df1 %>% group_by(town) %>% summarise(
p_males   = sum(gender == "m")/n(),
p_females = sum(gender == "f")/n(),
odds_males = p_males/p_females,
odds_females = p_females/p_males,
odds_ratio = odds_males/odds_females)