R编码：如何保留4个完整季度的数据记录

Question

I have a dataframe with company quarterly data and have this question:

How can I retain records for only those companies with 4 quarters of data (as companies sometimes appear with 1, 2 or 3 quarters of data but I need 4 complete quarters for each company across the entire dataframe)

I've included example R code below:

company<-c("xray", "xray", "xray",  "xray", "foxrot", "foxrot", "delta", "kilo", "kilo", "kilo", "kilo", "kilo", "kilo" )  

year <-c("1984","1984","1984","1984", "1985", "1985","1986", "1987","1988","1989","1989","1989","1989" )

qtr <-c("1","2","3","4", "1", "2","3", "4","1", "1","2","3","4")

IQ <- rnorm(13,0,10)  
REVQ <- rnorm(13,0,10)  
AssetQ <- rnorm(13,0,10)  
CashQ  <- rnorm(13,0,10)  

#Show dataframe  
data<-data.frame( year, qtr, company, IQ, REVQ, AssetQ, CashQ )

In this example 'xray' in 1984 and 'kilo' in 1989 should be the only companies remaining in the new dataframe. This example is unique in that you will notice the quarterly sequence 1-2-3-4 appears three times but only two of those sequences are good as the other (at rows 5-8) is a random occurrence. For the clean up to make sense each sequence 1-2-3-4 needs to be assigned to the same year and company.

This condition makes the task fairly tricky (at least for myself) and I've tried for nearly a day, searching the web and trying different methods, but nothing seems to work properly.

Thus, I'm kindly reaching out for some help.

Thank you~ M

Answer 1

Here is the code that can help you:

library(data.table)
data<-data.table( year, qtr, company, IQ, REVQ, AssetQ, CashQ)
fullyr <- data[,.(len=.N),by=.(year)][len == 4]
data <- data[year %in% fullyr$year]

Requesting you to provide the code you have tried for solution next time :)

Answer 2

The following code can help you ....

final=data.frame()
for(i in unique(data$company)){
temp=data[data$company==i,]
for(j in unique(temp$year)){
    if(nrow(temp[temp$year==j,])==4)
        final=rbind(final,data.frame(company=i,Year=j))
}
}

'final' dataframe will contain your required fields.

Answer 3

We can use data.table

library(data.table)
setDT(data)[data[, .I[uniqueN(qtr)==4], by = company]$V1]

Or

setDT(data)[, if(uniqueN(qtr)==4) .SD, by = company]
#   company year qtr         IQ       REVQ      AssetQ       CashQ
#1:    xray 1984   1  -5.827832   8.221870   9.6688477 -10.6321121
#2:    xray 1984   2   3.521643  -1.096940  -4.5014798  -0.9196087
#3:    xray 1984   3  -7.526160  -4.155428 -10.6556271   7.6872401
#4:    xray 1984   4  -7.255974   3.717738  -1.7913910   9.6325437
#5:    kilo 1989   1   2.252885 -19.238773   9.7476758   4.0115274
#6:    kilo 1989   2   9.018055 -12.411381  -0.3772812   6.8339812
#7:    kilo 1989   3 -12.221085 -13.040805   7.3529403   9.1510647
#8:    kilo 1989   4   2.088668  -7.753041   1.5701738 -11.2252986