R：如何用另一个数据框中的“ countif”值在数据框中创建新列？

Question

我有一个数据帧（df1），如下所示。 它表示公司活跃于特定市场的年份。

Company  Country  Year
A        Austria  2010
A        Germany  2010
A        Austria  2011
B        Italy    2010

现在，我有第二个数据帧（df2），如下所示。 它按投资类型列出给定时间公司在某个国家的所有投资，作为虚拟变量。

Company  Country  Year  JointVenture  M&A  Greenfield
A        Austria  2010  1             0    0
A        Austria  2010  0             1    0
A        Austria  2010  1             0    0
...

现在我的问题如下：我想在df1中添加一个新列 ，包括df2中指示的每种投资类型的“ countif” 。 例如，新的df1：

Company  Country  Year  Count.JointVenture  Count.M&A  Count.Greenfield
A        Austria  2010  2                   1          0
A        Germany  2010  ...........
A        Austria  2011
B        Italy    2010

另外，如何将新的列添加到df1，将这些计数转换为虚拟变量（如果> 0，则为1；如果0，则为0）？

感谢和抱歉，这是一个基本问题，但我在现有线程中找不到合适的解决方案。

马丁，干杯

Answer 1

使用aggregate（）和ifelse（）函数

# test data
df <- data.frame(Company = rep("A", 3), 
                 Country = rep("Austria", 3), 
                 Year = rep(2010, 3), 
                 JointVenture = c(1,0,1), 
                 MnA = c(0,1,0), 
                 Greenfield = rep(0,3))
# this is the new df
counts <- aggregate(cbind(JointVenture, MnA, Greenfield)~Country+Company+Year, data = df, FUN = sum)

# dummy
counts$dummyJointVenture <- ifelse(counts$JointVenture > 0, 1, 0)
counts$dummyMnA <- ifelse(counts$MnA > 0, 1, 0)
counts$dummyGreenfield <- ifelse(counts$Greenfield > 0, 1, 0)

Answer 2

我把data.table尝试扔进了竞技场：

df <- fread("Company  Country  Year
             A        Austria  2010
             A        Germany  2010
             A        Austria  2011
             B        Italy    2010")

df2 <- fread("Company  Country  Year  JointVenture  M&A  Greenfield
              A        Austria  2010  1             0    0
              A        Austria  2010  0             1    0
              A        Austria  2010  1             0    0")

setkey(df2, Company, Country, Year)
df2[,c("JointVenture", "M&A", "Greenfield") := .(sum(JointVenture), sum(`M&A`), sum(Greenfield)), by=.(Company, Country, Year)]
merge(x=df, y=unique(df2), by=c("Company", "Country", "Year"), all.x=T, all.y=F, suffixes = c("", "Count."))

导致

   Company Country Year JointVenture M&A Greenfield
1:       A Austria 2010            2   1          0
2:       A Austria 2011           NA  NA         NA
3:       A Germany 2010           NA  NA         NA
4:       B   Italy 2010           NA  NA         NA

Answer 3

使用dplyr::summarise_each并与Martin的数据merge 。

df <- fread("Company  Country  Year
             A        Austria  2010
             A        Germany  2010
             A        Austria  2011
             B        Italy    2010")

df2 <- fread("Company  Country  Year  JointVenture  MA  Greenfield
              A        Austria  2010  1             0    0
              A        Austria  2010  0             1    0
              A        Austria  2010  1             0    0")
library(dplyr)
df2 %>% 
  group_by(Company, Country, Year) %>%
summarise_each(funs(sum), JointVenture:Greenfield) %>%
full_join(df, by = c("Company", "Country", "Year")) -> df

编辑作业：更换了summarise与summarise_each从@zacdav输入和更换merge通过full_join留在dplyr