0

我制作了这段代码，通过平均时间来划分值，然后我输入了所需的初始模型。但我不明白。有人请给我任何建议吗？我想再次使用 reshape() 或 dcast() 吗？

AA<-read.table("AA_NYS_USD_1min_EST5EDT_200407s12e.dat", header=FALSE,as.is=TRUE, col.names=c("date","hour", "price","volume"))             
str(AA)
AA$date    <- as.Date(AA$date, "%d.%m.%Y")
AA$hour    <- format(strptime(AA$hour, "%H:%M:%S"), "%H:%M")
AA$houraid  <- as.numeric(as.factor(AA$hour)) 
head(AA, 3)
        data  hour price volume hourid
 1 01.07.2004 09:31 51.37  27900      1
 2 01.07.2004 09:32 51.32    100      2
 3 01.07.2004 09:33 51.45    700      3
require(reshape2)
library(reshape)
AA2 <- dcast(AA, data~hour, value.var='volume')   ### Utilizando texto das horas


head(AA2[,1:8], 3)
        date  09:31 09:32 09:33 09:34 09:35 09:36 09:37
 1 2004-07-01  27900   100   700  1000   800  3000  1600
 2 2004-07-02 145400   500    NA   500  6900    NA  4400
 3 2004-07-06   3200  2300  3800  1500    NA  1100  3700

library(dplyr)
a<-AA %>% group_by(hour) %>% summarise(mean = mean(volume,na.rm=TRUE))


for(k in 2:396){
  AA2[,k]<-AA2[,k]/a[k,2]
}
for( i in 1:nrow(a)){

  for(j in 1:nrow(AA)){

    if(AA[j,2]==a[i,1]){
      AA$volumeMean[j]<-AA[j,4]/a[i,2]

    }

  }

}
write.table(AA3,"AA_Complete.txt")

谢谢！！！

4

0 回答 0