R - 如何按组进行回归并获得预测值？

Question

personID<-c(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)
genger<-c('male', 'male', 'male', 'male', 'male', 'male', 'male', 'male', 'male', 'male', 'male', 'female', 'female', 'female', 'female', 'female', 'female', 'female', 'female', 'female', 'female', 'female', 'female', 'female', 'female', 'female')
height<-c(181, 161, 198, 195, 177, 175, 197, 195, 198, 193, 161, 167, 132, 181, 165, 151, 163, 180, 169, 181, 177, 135, 143, 107, 161, 142)
weight<-c(165,  73, 90, 89, 80, 159,    179,    177,    180,    175,    73, 76, 60, 165,    150,    69, 148,    164,    154,    165,    161,    61, 130,    97, 146,    65)
data<-data.frame(personID, genger, height, weight)
data

我是R初学者。

我喜欢按性别（男、女）进行回归。

回归公式为weight=solpe*height + intercept

我谷歌了一下，但我没看懂几篇文章。

我想要的输出如下。

person_id   gender  height  weight  predict_value  error
1            male   181      165       xxx           xx
2            male   161      73        ...           ...  
3            male   198      90 
4            male   195      89 
5            male   177      80 
6            male   175      159    
7            male   197      179    
8            male   195      177    
9            male   198      180    
10           male   193      175    
11           male   161       73    
12          female  167       76    
13          female  132       60    
14          female  181      165    
15          female  165      150    
16          female  151       69

如何按性别进行回归分析并添加预测和误差列？

如有任何帮助，我们将不胜感激。

Answer 1

这是一种方法。您可以拆分数据，执行回归并使用 predict() 找到置信区间，然后您可以将 return 拆分为原始结构。例如，使用您的测试数据并在样本数据

的 "genger" （原文如此）列上拆分

unsplit(lapply(split(data, data$genger), function(x) {
    m<-lm(weight~height, x)
    cbind(x, predict(m, interval ="confidence"))
}), data$genger)

这个returns

   personID genger height weight       fit        lwr       upr
1         1   male    181    165 124.17126  94.106766 154.23576
2         2   male    161     73  87.11321  29.280886 144.94554
3         3   male    198     90 155.67061 115.126629 196.21458
4         4   male    195     89 150.11190 113.707198 186.51660
5         5   male    177     80 116.75965  83.508504 150.01080
# etc...

R - 如何按组进行回归并获得预测值？

R - How to do regression by group and getting predict values?

r

statistics