Question

我有下表的胃灼热症状和对两种不同药物的反应：

           Medication
Symptoms    Drug A Drug B Totals
  Heartburn     64     92    156
  Normal       114     98    212
  Totals       178    190    368

我正在寻找 R 功能，该功能通过以下操作来获取预期的计数：

                    Medication
Symptoms       Drug A                   Drug B               
Heartburn     156 * 178 / 368 = 75      156 * 190 / 368 = 81 
Normal        212 * 178 / 368 = 103     212 * 190 / 368 = 109

理想情况下，显示器甚至可选择如下：

           Medication
Symptoms       Drug A   Drug B    Totals
  Heartburn     64(75)   92(81)    156
  Normal       114(103)  98(109)   212
  Totals       178      190        368

问题不是太多，不是吗？

Answer 1

来自?chisq.test的示例：

 M <- as.table(rbind(c(762, 327, 468), c(484, 239, 477)))
 dimnames(M) <- list(gender = c("F", "M"),
                     party = c("Democrat","Independent", "Republican"))
 Xsq <- chisq.test(M)  # Prints test summary

我认为结果的$expected组件是您想要的：

 Xsq$expected
 ##       party
 ## gender Democrat Independent Republican
 ##      F 703.6714    319.6453   533.6834
 ##      M 542.3286    246.3547   411.3166

首选显示的开头：

M2 <- M; 
M2[] <- paste(M,paste0("(",round(Xsq$expected),")"))
##      party
## gender Democrat  Independent Republican
##      F 762 (704) 327 (320)   468 (534) 
##      M 484 (542) 239 (246)   477 (411)

另见?addmargins

Answer 2

以下是一些替代方案：

1）外部使用outer生成预期值，然后使用sprintf将其与原始表和parens放在一起。 noquote可用于显示它而不带引号：

mm <- m
mm[-3,-3] <- matrix(sprintf("%3.0f(%.0f)", m, outer(m[, 3], m[3, ]) / m[3,3]), 3)[-3,-3]

，并提供：

> noquote(mm)
           Medication
Symptoms    Drug A   Drug B   Total
  Heartburn  64(75)   92(81)  156  
  Normal    114(103)  98(109) 212  
  Totals    178      190      368

2）gmodels :: CrossTable 此函数专门用于生成包含预期计数和其他统计信息的交叉表：

library(gmodels)
CrossTable(m[-3, -3], expected = TRUE, prop.r = FALSE, prop.c = FALSE, 
                        prop.t = FALSE, prop.chisq = FALSE)

，并提供：

   Cell Contents
|-------------------------|
|                       N |
|              Expected N |
|-------------------------|


Total Observations in Table:  368 


             | Medication 
    Symptoms |    Drug A |    Drug B | Row Total | 
-------------|-----------|-----------|-----------|
   Heartburn |        64 |        92 |       156 | 
             |    75.457 |    80.543 |           | 
-------------|-----------|-----------|-----------|
      Normal |       114 |        98 |       212 | 
             |   102.543 |   109.457 |           | 
-------------|-----------|-----------|-----------|
Column Total |       178 |       190 |       368 | 
-------------|-----------|-----------|-----------|


Statistics for All Table Factors


Pearson's Chi-squared test 
------------------------------------------------------------
Chi^2 =  5.85     d.f. =  1     p =  0.0156 

Pearson's Chi-squared test with Yates' continuity correction 
------------------------------------------------------------
Chi^2 =  5.35     d.f. =  1     p =  0.0207

3）descr :: CrossTable 这与gmodels :: CrossTable类似，但输出略有不同。

library(descr)
CrossTable(m[-3,-3], prop.r = FALSE, prop.c = FALSE, prop.t = FALSE,
    prop.chisq = FALSE, expected = TRUE)

，并提供：

   Cell Contents 
|-------------------------|
|                       N | 
|              Expected N | 
|-------------------------|

====================================
             Medication
Symptoms     Drug A   Drug B   Total
------------------------------------
Heartburn        64       92     156
               75.5     80.5        
------------------------------------
Normal          114       98     212
              102.5    109.5        
------------------------------------
Total           178      190     368
====================================

注意：我们将此作为输入m：

m <- matrix(c(64, 114, 178, 92, 98, 190, 156, 212, 368), 3, 
  dimnames = list(Symptoms = c("Heartburn", "Normal", "Totals"),
                  Medication = c("Drug A", "Drug B", "Total")))

是否有R函数来获取预期计数表？

2 个答案: