簡體   English   中英

在R中重新記錄時間序列數據

[英]recoding time series data in r

我正在嘗試使用超時結構來重新編碼現有數據。 我的數據集如下所示:

dput(z)

structure(list(democracy = c(0L, 0L, 0L, 0L, 0L, 0L, 1L, 1L, 
1L, 0L, 0L, 0L, 0L, 0L, 0L, 0L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 
1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 
1L, 1L), year.x = 1967:2008, time = c(1, 2, 3, 4, 5, 6, 7, 8, 
9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 
25, 26, 27, 28, 29, 30, 31, 32, 33, 34, 35, 36, 37, 38, 39, 40, 
41, 42)), .Names = c("democracy", "year.x", "time"), row.names = 176:217, class = "data.frame")

因此,我想創建一個新的變量,例如time.democ,如果democracy==0 ,則取值為零,但如果從democracy ==1 ,則從1開始,直到democracy==0 ,然后重新開始計算時間段。再次。 我將在一系列國家/地區進行此操作,但是我假設一旦正確使用此功能,使用ddply進行泛化就足夠容易了。 有什么建議么?

我想得到這個:

dput(z)

structure(list(democracy = c(0L, 0L, 0L, 0L, 0L, 0L, 1L, 1L, 
1L, 0L, 0L, 0L, 0L, 0L, 0L, 0L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 
1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 
1L, 1L), year.x = 1967:2008, time = c(1, 2, 3, 4, 5, 6, 7, 8, 
9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 
25, 26, 27, 28, 29, 30, 31, 32, 33, 34, 35, 36, 37, 38, 39, 40, 
41, 42), new.time = c(0, 0, 0, 0, 0, 0, 0, 1, 2, 0, 0, 0, 0, 
0, 0, 0, 0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 
16, 17, 18, 19, 20, 21, 22, 23, 24, 25)), .Names = c("democracy", 
"year.x", "time", "new.time"), row.names = 176:217, class = "data.frame")

謝謝!

您可以結合使用rlesequence來執行此操作。 rle執行游程長度編碼,而sequence生成序列。

z$new.time <- sequence(rle(z$democracy)$lengths)
z$new.time[z$democracy==0] <- 0

head(z, 20)

    democracy year.x time new.time
176         0   1967    1        0
177         0   1968    2        0
178         0   1969    3        0
179         0   1970    4        0
180         0   1971    5        0
181         0   1972    6        0
182         1   1973    7        1
183         1   1974    8        2
184         1   1975    9        3
185         0   1976   10        0
186         0   1977   11        0
187         0   1978   12        0
188         0   1979   13        0
189         0   1980   14        0
190         0   1981   15        0
191         0   1982   16        0
192         1   1983   17        1
193         1   1984   18        2
194         1   1985   19        3
195         1   1986   20        4

多謝您的回覆。 我遵循了您的建議,最后編寫了一個函數,以便可以通過ddply將其應用於我的(縱向)數據集中的所有單元。 我發布它是因為它可能會幫助其他人,盡管我確信還有更優雅的解決方案:

# is a long format data frame
new.time <- function(a){
    a <- a[order(a$year.x),]
    a$new.time <- sequence(rle(a$democracy)$lengths)-1
    a$new.time[a$democracy==0] <- 0
    return(a)
}

merged1 <- ddply(merged, .(country.x), new.time)

暫無
暫無

聲明:本站的技術帖子網頁,遵循CC BY-SA 4.0協議,如果您需要轉載,請注明本站網址或者原文地址。任何問題請咨詢:yoyou2525@163.com.

 
粵ICP備18138465號  © 2020-2024 STACKOOM.COM