Question

我有数据框，想要重新编码我可以比较未来实验的日期。但它不能用if语句重新编码日期。也许有提示这样做。

test <- data.frame(StartDate = c(rep("2015-02-16", 5), rep("2015-02-17", 5), rep("2015-02-23", 5), rep("2015-02-24", 5)),
                   response = sample(1:10,20,rep=TRUE)*7)



test$StartDate <- strptime(test$StartDate, format="%F")

test$day <- test$StartDate$mday
test$experimentday <- ifelse( (test$day == 16), 1, NA)
test$experimentday <- ifelse( (test$day == 17), 2, NA)
test$experimentday <- ifelse( (test$day == 23), 1, NA)
test$experimentday <- ifelse( (test$day == 24), 2, NA)

    StartDate response day experimentday
1  2015-02-16       35  16            NA
2  2015-02-16       35  16            NA
3  2015-02-16       63  16            NA
4  2015-02-16       63  16            NA
5  2015-02-16       21  16            NA
6  2015-02-17       70  17            NA
7  2015-02-17       42  17            NA
8  2015-02-17       14  17            NA
9  2015-02-17       42  17            NA
10 2015-02-17       70  17            NA
11 2015-02-23       49  23            NA
12 2015-02-23       21  23            NA
13 2015-02-23       42  23            NA
14 2015-02-23       14  23            NA
15 2015-02-23       21  23            NA
16 2015-02-24       56  24             2
17 2015-02-24       42  24             2
18 2015-02-24       42  24             2
19 2015-02-24       21  24             2
20 2015-02-24        7  24             2

其他变体也不起作用

ifelse( test$day == 16, test$experimentday==1, test$experimentday==NA)

if(test$day == 16) {test$experimentday2 <- 1} 
else {test$experimentday2 <- NA}

if(test$day == 16) test$experimentday3 <- 1 
else test$experimentday3 <- NA

Answer 1

您的代码正在按照您的说法进行操作。代码的最后一行，

test$experimentday <- ifelse( (test$day == 24), 2, NA)

如果experimentday为24，则

将day的值设置为2，并将experimentday的所有其他值设置为NA，从而有效地撤消前一行代码的工作。你可能想尝试这样的事情，而不是：

oldvalues <- c(16, 17, 23, 24)
newvalues <- c(1, 2, 1, 2)
test$experimentday <- newvalues[match(test$day, oldvalues)]

如果其他声明不适合约会

1 个答案: