Pregunta

I have two files which contain time series in this format:

Tag, T1, T2, ...  Tn
Stock-1, 2, 3, ..  6
Stock-2, 3, 2, ..  7
.
.
Stock-m, 2, 1, ..  9

In R, for any tag (ex. Stock-x) I want to take related arrays from both files and calculate cross correlation(ccf) for them. How I can do that ?

Note: If it would be easier, I can change the file format.

Update:

I read from file into data.frames with this:

file1 = read.table("file1.txt", header = TRUE, sep = ",")

How will I reach to specific rows with Tag name such as "Stock-5" ?

¿Fue útil?

Solución

I removed the "Tag" from the files:

T1, T2, ...  Tn
Stock-1, 2, 3, ..  6
Stock-2, 3, 2, ..  7
.
.
Stock-m, 2, 1, ..  9

Then I was able get results with this R code:

file1 = read.table("file1.txt",head=TRUE,sep=",")
file2 = read.table("file2.txt",head=TRUE,sep=",")

row1 = file1["Stock-5",]
row2 = file2["Stock-5",]

vec1 = c(t(row1))
vec2 = c(t(row2))

ccf(vec1,vec2)

Otros consejos

It seems that the question might be answered with:

f1stock5df <- subset(file1,Tag %in% c("stock5"))

f2stock5df <- subset(file2,Tag %in% c("stock5"))

bothstocks <- rbind(f1stock5df, f2stock5df)

with(bothstocks, ccf(x,y))
Licenciado bajo: CC-BY-SA con atribución
No afiliado a StackOverflow
scroll top