标量乘以r中的两个data.frames答案

【问题标题】：scalar multipling two data.frames in r标量乘以r中的两个data.frames
【发布时间】：2021-04-19 10:47:53
【问题描述】：

我在 r 中有两个 data.frames。第一个是

                cases      population  urbanisation  density temperature  h_dev_index
Austria         563.375758 10.7969091     63.07388  134.08690   13.011172   0.9898000
Belgium         109.400000 13.5885455     99.38933  443.52112   16.185297   0.9829455
Bulgaria          0.000000  5.5320606     85.84782   52.30011   20.825068   0.9669576
Croatia           2.000000  3.4548485     64.21382   60.46082   20.855372   0.9288667

第二个是：

估计标准。误差t值Pr(>|t|)

(Intercept)                        -1.916e+00  7.144e-01  -2.682  0.00785 ** 
population                         -7.327e-03  1.572e-03  -4.659 5.37e-06 ***
cases                               1.473e-03  1.544e-04   9.541  < 2e-16 ***
urbanisation                        3.798e-03  1.962e-03   1.936  0.05410 .  
density                             8.132e-05  2.512e-04   0.324  0.74641    
temperature                        -3.518e-02  8.641e-03  -4.071 6.43e-05 ***
h_dev_index                         1.842e+00  8.800e-01   2.093  0.03743 *

我需要将第二个数据帧的第一列的系数乘以数据帧。结果应该是（对于奥地利）：

-1.916e+00 + 563.375758*(-7.327e-03)+ 10.7969091*(1.473e-03)+ 63.07388*(3.798e-03)+134.08690*(8.132e-05)+     13.011172*(-3.518e-02)+   0.9898000*1.842e+00

我应该如何解决这个问题？

【问题讨论】：

不应该是10.7969091*(-7.327e-03)之类的吗？
是的，适用于所有行
我添加了一个可能的解决方案。希望有帮助。请注意数据的名称。

标签： r dataframe vector multiplication

【解决方案1】：

尝试重塑和合并：

library(tidyverse)
#Code
res <- df1 %>% rownames_to_column('id') %>%
  pivot_longer(-id) %>%
  left_join(df2 %>% select(1,2) %>% rename(name=V1)) %>%
  group_by(id) %>%
  summarise(Val=sum(value*V2)+df2$V2[df2$V1=='(Intercept)'])

输出：

# A tibble: 4 x 2
  id          Val
  <chr>     <dbl>
1 Austria   0.451
2 Belgium  -0.200
3 Bulgaria -0.578
4 Croatia  -0.712

使用的一些数据：

#Data
df1 <- structure(list(cases = c(563.375758, 109.4, 0, 2), population = c(10.7969091, 
13.5885455, 5.5320606, 3.4548485), urbanisation = c(63.07388, 
99.38933, 85.84782, 64.21382), density = c(134.0869, 443.52112, 
52.30011, 60.46082), temperature = c(13.011172, 16.185297, 20.825068, 
20.855372), h_dev_index = c(0.9898, 0.9829455, 0.9669576, 0.9288667
)), class = "data.frame", row.names = c("Austria", "Belgium", 
"Bulgaria", "Croatia"))

df2 <- structure(list(V1 = c("(Intercept)", "population", "cases", "urbanisation", 
"density", "temperature", "h_dev_index"), V2 = c(-1.916, -0.007327, 
0.001473, 0.003798, 8.132e-05, -0.03518, 1.842), V3 = c(0.7144, 
0.001572, 0.0001544, 0.001962, 0.0002512, 0.008641, 0.88), V4 = c(-2.682, 
-4.659, 9.541, 1.936, 0.324, -4.071, 2.093), V5 = c("0.00785", 
"5.37e-06", "<", "0.05410", "0.74641", "6.43e-05", "0.03743"), 
    V6 = c("**", "***", "2e-16", ".", "", "***", "*"), V7 = c("", 
    "", "***", "", "", "", "")), class = "data.frame", row.names = c(NA, 
-7L))

请注意数据框的名称。

【讨论】：